Register for free to join our community of investors and share your ideas. You will also get access to streaming quotes, interactive charts, trades, portfolio, live options flow and more tools.
Register for free to join our community of investors and share your ideas. You will also get access to streaming quotes, interactive charts, trades, portfolio, live options flow and more tools.
CJ, really? If you mean work out as in goes well, congrats! If you mean work out as in becomes reality, good luck!
Paul
wbmw, re: "I'm guessing they only have NDA versions, because they don't want Opteron power dissipation to be compared to Intel's."
Why would you assume that? All anecdotal evidence so far, even with overclocked Opterons and A64s, has shown that they run very cool(relative to XP and P4, at least). I'll grant you there may be something they're trying to hide, but I don't think PD is it.
Paul
wbmw - get real. WHICH current Opteron is an 84W chip? Be specific, please.
chipguy, ask some of the guys around here if they could use a 64bit Opteron notebook in their work, instead of being tied to their workstation. I don't remember specifically who mentioned it a while back, maybe yourbankruptcy? sgolds? Combjelly?
Paul
Sun likes Solaris on Opteron - "phenomenal"
Paul
"We are seeing phenomenal performance on Opteron. We are running Solaris on Opteron in our labs," Jonathan Schwartz, executive vice president of software at Sun, said in a recent interview. "Intel and AMD are really important" to the future of Sun, he added.
http://zdnet.com.com/2100-1103_2-5077111.html?tag=zdnnfd.main
wbmw, it is also possible that they're confident enough in their process now to be able to only drop the voltage and keep the frequencies @ present desktop levels. Which, of course, would mean the desktop frequencies would be climbing. One can hope, no? ;)
Paul
Really, EP? What 30W 64bit CPU does Intel have in the works?
Paul
EP - MVS? Your thought may be valid, but the MS guy had no reason to assume that, since the questioner didn't give any reason to believe he was doing that.
Paul
First image I've seen of real world topless K8. It's a mobile. :) Unfortunately, they didn't clean off the heat transfer grease, so you can't see if there are any bridges a la Athlon XP. Interesting that the screenshots and BIOS show 1.8 GHz 3200, but the chip shown is a 3000. Maybe an old BIOS.
Paul
http://forums.tweaktown.com/showthread.php?s=&threadid=13305
UpNDown - ? No reason? How about unstable software? Just a thought. Besides, this isn't an official press release or anything, just an MS guy responding in a forum.
Paul
UpNDown, sounds like FUD to me. The link is incomplete, by the way.
Paul
Interesting quote from the tbreak.com A64 thread:
"We turned off the fan for a couple of mintutes. It hit around 45C max and that's after some time. It runs at 39C max even in hot room while benching at its max overclocked speeds. I can assure you that heatsink didn't even get warm. again, i can assure you that chip runs way cooler than P4, ask Abbas and don't tell me he's biased towards any brand."
http://www.tbreak.com/forums/showthread.php?s=a91c680b99c5adf638f7d84bbd0a21bd&threadid=19317&am...
sgolds, yes, that would be a nice exchange!
Keith, (finally found OPN ID!)according to this link, they're wrong. I was referring to another 64FX(at tbreak.com), but the one at Akiba is also 1.50V.
Paul
http://homepage3.nifty.com/aona/opn/opn_amdk8.html
I'm going to keep a copy on my website and post a link to it on the yahoo amd-intel info site.
sgolds, I think there may be more OCing enthusiasts than you imagine. I see multitudes of them on many of the boards I visit. Some of us are here, for that matter. OCing won't likely go away, it'll just take more effort and ingenuity to do it. Haven't you ever liked getting the most for your money? I'm on a tight budget, so when I can take a $40 XP1700+ and get the performance of a 2400+ or 2600+ without radically overvolting, and up bus speeds to get a lot more bang for the buck, I do it. There's just something fun about tweaking, as well, kind of the modern day equivalent of the guy in his garage building his hot rod. Now if it gets to the point where it takes too much effort for the return, it may fade away; but we're certainly nowhere near that now. I do share your concerns about unethical vendors ripping off their customers, though, and when I hear of them, I don't hesitate to spread the word.
Paul
CJ & Jozef, it's still there. ADAFX51CEP5AK is the part #, so I think you said that meant 1.5V and 70C case, CJ.
Paul
http://www.tbreak.com/nasser/Naser-A64/3.gif
A64 & 64FX to be compared soon. These guys have good sources, don't they? As well as no NDAs.
Paul
http://www.tbreak.com/forums/showthread.php?s=&threadid=19378
son of thera, right you are, it was early when I read it! ;)
Makes it even more impressive, doesn't it?
Paul
64FX 2.2GHz teaser
http://www.hardwareanalysis.com/content/article/1649/
avatar, 24182 wasn't their best score, either. Scrolling down, you will find a score of 24825 @ 2390/239"FSB"/DDR333setting(@196MHz). RAM couldn't handle 220MHz+. Quake score was 587fps. Impressive. This was with a Via chipset and plain water cooling, no refrigeration.
Paul
aceshardware thread discussing it and links to Futuremark thread as well.
http://www.aceshardware.com/forum?read=105035305
wbmw, I can buy that.
Paul
Golfbum, if I understand DDB correctly, the optimizations are in hardware, so yes, they would be unique to AMD. But developers won't have to optimize for it, it's a hardware implementation of optimizations in how the processor handles SSE2 instructions as they exist, so it sounds like it would be transparent to the developer.
Paul
DDB, oops, you included the parenthesis in the URL.
DDB, oops, you included the parenthesis in the URL.
gollem, digit-life must have posted this by mistake, since they've pulled the article, so I don't have the image, but got to this post and you'll see the specs. 16GB.
http://www.investorshub.com/boards/read_msg.asp?message_id=1398893
Edit:
found it elsewhere!
http://www.search4hardware.com/2/p_2_1386_Iwill_DK8X.html
larger image
http://www.overclockers.co.nz/news/images/iwill_dk8x.jpg
NaS, can't say I've ever experienced a problem from either PM or DI, but I always get the OEM Windows anyhow, so support isn't an issue.
NaS, thanks, but I remembered that I could do it in a roundabout way with Drive Image 5 (do image, delete partitions, restore image and resize while restoring). So it's done and all is well.
Paul
Thanks, Petz, I missed all that. Looks encouraging!
Paul
Thanks for the clarification, I'd forgotten exactly how the memory clock was derived.
Where did all this talk about SSE2 improvements come from? I was out of the loop from the 23rd through the 28th, was this revealed then? Who made the claim?
Paul
OT - does Win2K have any sort of utility for resizing partitions, or will I need to use Partition Magic or a similar program? I just started using Win2K and found out that my old Partition Magic 4 won't support its version of NTFS. TIA,
Paul
tecchannel will have a 2GHz Opteron workstation review Sep 10th.
am 10..09.: AMDs Opteron mit 2 GHz im Workstation-Test
AMD bereitet sich auf den Launch des Athlon64 vor und schraubt den Takt nach oben. Wir testen die neue CPU in einem Board, das auch für den kommenden Mainstream-Prozessor geeignet ist.
Rough translation:
on Sep 10th: AMD's Opteron @ 2GHz in Workstation test. AMD gets ready for the Athlon 64 launch and cranks up the clockspeed. We test the new CPU in a board that will also support the upcoming mainstream processor.
wbmw, what I'm asking is exactly how the memory frequency is affected. Since we know that is doesn't always change linearly as the multiplier changes, what do we actually know about how it changes when the external(HT based)clock changes? Yes, we know the memory controller frequency will be the same as the CPU clock, but will the memory frequency merely be the expected multiple(e.g. 222/200) of whatever the standard memory frequency for that memory type setting(DDR200/266/333) is, or does it change in some other fashion? Could it perhaps change increments based on a HT "bus" speed range, for example? I wouldn't really expect it does, but since I haven't seen anyone specifically list the memory frequencies in their reviews, I wouldn't state categorically one way or the other. I wonder if this is one of the bits of info involved in AMD's NDA that Magee was talking about.
Paul
chipguy, there are still components on the MB and trace routing decisions that can affect stability and performance.
Paul
This is a pretty serious MB. Iwill has really concentrated on Intel for server and workstation MBs until now.
Paul
http://www.digit-life.com/news.html?95341#95341
DK8X main specs:
* AMD Opteron (Socket 940) CPU
* AMD 8111 + 8131 + 8151 chipset
* Memory: up to 16Gb dual-channel DDR333/266/200 w/ECC, 8 x DIMM sockets
* AGP Pro/8X slot
* 2 x PCI-X (64bit/133MHz), 1 x PCI (64bit/66MHz), 2 x PCI (32bit/33MHz)
* Built-in 1000Mbps Ethernet: 3COM 3C940
* Serial ATA controller: Silicon Image Sil3114, 4 x Serial ATA ports supporting RAID 0, 1, 0+1
* 2 x Ultra DMA 133/100/66/33 channels
* TI TSB43AB22 IEEE1394 controller
* NEC USB2.0 controller
* eATX form-factor (12" x 13" or 304x330mm)
* Power: EPS 12V (460W)
Keith, this from the preview is very interesting.
http://www.anandtech.com/cpu/showdoc.html?i=1856&p=8
I'm not at all sure I agree with the reasoning earlier in the article about the HT clock not affecting results.
So, for our 1.8GHz Opteron 144, the multiplier is 9x, which is why raising the HT frequency to 222MHz increases the clock speed to around 2GHz. But we are increasing the HyperTransport clock in our overclocking, and not a FSB clock, which does not exist on Opteron/Athlon64. In real terms, this means our CPU overclocking has a significant impact on Performance, but it is unlikely that our increase in memory speed will have nearly as much impact on performance. Since we are nowhere near saturating the Hypertransport bus at 200 (effective 800), increasing HyperTransport to 222 (888) will not likely have much, if any, impact on overall performance.
Any thoughts? Do we actually know what the memory frequency is at that "FSB"? I don't see where they list it. They used 9x multiplier and 222MHz "FSB". I know what it would be at 9x and 200 "FSB"(DDR200 = 200, DDR266 = 257.14, DDR333 = 327.26), but does it just increment in a linear manner with the HT overclock?
Paul
Opteron 248 headed for the light
http://www.theinquirer.net/?article=11371
wbmw, it's not a matter of what AMD has, but more of the cost of memory. When it gets down to where average serious gamers generally can afford those levels of memory, then the game developers will start putting out games that will make use of it.
That's probably why Tim Sweeney says that 64bit developer rigs are pretty much necessary for them now.
Paul
kpf, that's not usually how MB makers list it. Normally it would mean that 2GB total is what is supported.
Paul
wbmw, obviously it doesn't, but it's an older chipset/board design that still may have some issues. In any case, the main arena in which the greater memory capacity will be important will be in workstations and servers. Eventually high-end gaming will be wanting larger memory, but 2GB is enough there for now.
Paul
First A64 "review". ALi(yuck) reference board. M1687 chipset has been ready for quite a while, lacks features. Eval BIOS and CPU(was labeled 3100+, changed to 2900+, reflected in other board's(hmm..} BIOS. OC'd from 200 to 218 "FSB" / 1800 to 1965MHz CPU.
Paul
http://www.ocworkbench.com/ocwbcgi/newspro/viewnews.cgi?newsid1062431028,75743,