Printable View
Yeah there is; $12.00, different packaging, and the argument they're binned higher.. :rolleyes: Why would that be? Who OC's a server? The worst 6600 on the planet OC's 1 Gig by accident. I must be missing the point... Whatever, have a good one...Quote:
Originally Posted by dinos22
EDIT: Sorry for going OT.
Definitely looking forward to seeing some of your benches. I have a couple of 7900GT's that run @ 7900GTX specs so I should be able to acheive your non overclocked results.Quote:
Originally Posted by bingo13
this is the same argument over San Diego and Opteron what can i say.....if you haven't been following then you probably won't know better nowQuote:
Originally Posted by pcdoc1
Not binned in terms of higher speeds, but binned in terms of higher quality and durability. They are tested for higher temperatures at lower voltages, which in turn (as we learned from mobile athlons and 939 opterons), leads to higher overclockability.Quote:
Originally Posted by pcdoc1
Thanks.Quote:
Originally Posted by bingo13
Why do they keep connecting them over PCI? It's a shame this board is soo cheap on PCI-E lanes...
Could you please do a Raid5 test anyway? :buddies: I'm still hoping that I don't have to buy a proper raid card.
Bull, they are binned at stock speed and vcore... Please tell me how you measure for higher quality when your worst product has 40% headroom...They get great yields and just like AMD will soon start OCing they own chips to a higher retail rating... This dog doesn't hunt...Quote:
Originally Posted by EnJoY
Once again we are OT so I suggest we get back to the topic at hand, 6 month old vaporware...
I THINK the xeons are also architecturaly different. 1 is better at number crunching and another is better at what ever. i think if thats what i heard right.Quote:
Originally Posted by pcdoc1
Duhhh, you heard wrong...Quote:
Originally Posted by tvdang7
Quote:
Originally Posted by pcdoc1
I didn't state that as my opinion, I stated that as fact. Don't let your ignorance disable your ability to learn. There is nothing else to discuss on this, it's just what it is.
well chill out dude. i said i think . i suggest u go back to not posting again.:woot:Quote:
Originally Posted by pcdoc1
And our opinions are based on results being posted the last couple weeks. The xeon thread has results of 3060s priming at 3.7ghz and above when there are still recent week 6600s that wont even boot 3.7ghz.
http://www.xtremesystems.org/forums/...d.php?t=122224
I think we all understand that we shouldnt whine that our 2.4ghz cpu wont boot and run orthos at 3.7ghz. But some of us want to eek out all the performance possible out of our $300 cpus and the xeons are looking good right now. I've yet to see someone complain of a dud maybe only sky high temps. I'm not trying to say intel sprinkled the xeons with pixie dust just that is it that unbelievable that they may just be a tad bit better.
Back Ot
This board is looking really decent. I cant wait to see some solid 9x and 10x multi fsb results.
Yup. I LOVE my Xeon. Clocks high with no to little voltage increase. The Xeon with the BadAxe2 is a great combo so far. I also have a DFI CFX3200 that kinda leaves a bad taste in my mouth. Don't get me wrong, I had no real issues with mine and it overclocks like a champ, but the utter disrespect that DFI gave to us owners is hard to forgive. :nono:
Back on topic...I guess that's probably why I won't buy this DFI although I'm sure that it will be fast.
any way we could come back to the topic?
Guys, any more OT I will close the thread. Take the xeon discussion to another thread please.
Is the board picky for memory like other LP boards have been?
Tony, any of the Cold Boot problems that have plauged previous DFI boards?
i didn't know DFI had cold boot problems with any other Intel based mobos :confused:Quote:
Originally Posted by qdemn7
I wasn't refering to Intel boards, just DFI boards in general. EDIT see here:Quote:
Originally Posted by dinos22
Quote:
Technical issues
Recently, many "cold boot" issues have been determined, relating to various issues in the power supply. Additionally, there have been no reproducible instances of CPU damage outside of user error.
DFI offers a wide range of motherboards with the ability to configure the memory timings well beyond what's normally available to users of other motherboards as well as countless other settings for changing things such as the speed and voltages of various parts. This allows for out of warranty overclocking to the point of several world records. The boards are designed with tight tolerances to provide this overclocking ability which has earned them the reputation of being picky about the components selected.
The DFI Infinity series is well known for having "cold boot" issues, with many people encountering this problem.
More recently, the DFI LanParty RDX-200 CF-DR has been having various functional issues with ATI's ATI CrossFire technology as well as non-working CPUs, from bugs in the BIOS code. Many of the issues found in the RDX-200 CF-DR have been fixed in the board's successor, the Lanparty CFX3200-DR UT.
DFI boards are generally considered to be 'fussy' about the types of memory used in them, preferring Samsung (TCCD, TCC5, TCC3, UCCC) or Winbond (BH5 UTT) chips over 'generic' branded RAM. DFI boards are also known to have issues with Crucial's Ballistix line of memory (using Micron memory), with hundreds of cases being reported of dead ram.
Well for those of us in the Netherlands there wont be a RD600 till end of January. I called their office in Rotterdam and a rep said that due to shipping schedules and logistics of other kind there will be no DFIs in the Benelux till the end of january :cool:
Tony does rd600 support ECC?
well here (France), there's a shop saying (29th november) that dfi rd600 will be available in 15 daysQuote:
Originally Posted by Ibinco
just a few days to see if it was true ;)
Quote:
Originally Posted by metalazzo
Url to the shop please ;)
Quote:
Originally Posted by metalazzo
yes linkage!!!!
Thats just some wishful thinking on that French E-tailers part. I wouldn't take it seriously gents.
DFI Support Europe.
Wow! 235 Euros, that's cheap! I thought that the price would be atleast 300 Euros here in Europe. I don't think they will get the board but atleast we got some kind of idea of the price here in europe. The Asus Striker Extreme is 330 Euros here in Sweden.Quote:
Originally Posted by metalazzo
I hope Tony has heard something more exact about the number of boards and a date for europe.
I agree with you and Bulldog : they won't have the board before 2007 :)Quote:
Originally Posted by Gambit_2K
but I don't think 235€ is cheap... I think 680i crappy mb prices are really foolish and it must not hide the fact that mb prices are skyrocketing :/
When did anandtech get their review board meaning should I be refreshing their page every 10 seconds? :D or is it likely we have to wait a few days until they are done with the article?
Yeah your right, just a year ago I would never ever even thought of buying a a board for over 200 Euros. I wonder if there is some sort of Cartel building?Quote:
Originally Posted by metalazzo
That French E-tailer is way off base or his info is pretty outdated. The only 680i chipset boards we have planned are still on the drawing board. Nothing close to a production model yet.
DFI Support Europe.
But hmm didn't Tony mention that he had one with him from his trip? How early is that from an actual productionmodel then?Quote:
Originally Posted by Bulldog14
this french e-tailer didn't talk about a 680i dfi :nono:Quote:
Originally Posted by Bulldog14
just about dfi rd600 and some 680i based mb (and no dfi :stick: )
My French is pretty bad, but indeed if you look further into the details he does make mention that the 680 board refers to an Asus model. The way it's setup on the main page is a bit deceptive however.
Hi guys! I´ve a simple question about this board. Is possible to increase vcore and keep EIST and/or C1E activated? I come from a P5W and this is not possible, I know other mobos can do it, but this one?
PS: Thanks Tony for all the info :toast:
hehe you don't want to know how long R&D has been working on the RD600 board. Oskar and his team are perfectionists ;) so I'm not expecting a 680 board to come across my desk anytime soon. Priorities can change overnight concerning MB R&D, but realistically I don't see a production model anytime soon.
Just passed Orthos at four hours with these settings, still have a lot of tuning left to do. Too many tests, not enough time. :D
http://img295.imageshack.us/img295/8...007x511iw9.png
that is sweet ..!!!! :banana:Quote:
Originally Posted by bingo13
really appreciate the your guys effort in testing the MB :clap:
looking good.. the cpuz shot of SB600 reminds me... wonder how the SATA raid controller performs on this board for raid 0 etc... guess something for much later to test huh :)Quote:
Originally Posted by bingo13
It performs better than the on-board Promise. ;) I have gone through about seven drive images in RAID testing so far. Of course I am testing at 7x511 as a starting point to figure out at what settings RAID is stable.Quote:
Originally Posted by eva2000
My personal viewpoint is that the "motherboard" RAID solutions are just not worth it, sure you get nice HDTach/PCMark05 results, but real application results for the most part are barely faster if not sometimes slower. I am testing it anyway. :)
ah but but... us pcmark05 junkies like fast hdds :DQuote:
Originally Posted by bingo13
Sorry, don't pay any attention to my post, I was just mad about some failed 680i boards and was taking it out on this poor little newcomer. ;)Quote:
Originally Posted by Tony
Haha, hadn't read any RGonian since the PCPer days, always a pleasure :toast:Quote:
Originally Posted by RGone
No support for ECC . You can also read it in the manual .Quote:
Originally Posted by deadlock7
bingo13, is that because your RAID got corrupted after locks/reboot as a cause of experiment with OC, OR you see SATA/RAID errors/lockups while running HDD untils (like with current 680i mobos)?Quote:
Originally Posted by bingo13
Thanks!
Nice clocking on the FlexXLC there. Was this with 2.3V or lower perhaps?Quote:
Originally Posted by bingo13
Now show us 550FSB! :stick:
I doubt you will see 550fsb from this board, 511 max in bios and 520 flakey stable in XP.Quote:
Originally Posted by pcmoddingmy
testing raptors in raid 0 on ati raid now ;)
Sweet! :woot:Quote:
Originally Posted by Tony
it's gonna be hard for 6600 users on p/c :(Quote:
Originally Posted by Tony
6600 uses a 9 multi as far as I know, 9x500 is 4.5GHZ, is that not enough?
sweet. i'll wait untill i know if i want crossfire or SLI again.
ps i have a 3060 xeon. and it was £50 cheaper than a E6600 would have cost me
:woot:
Tony you know the saying, it's never enough :DQuote:
Originally Posted by Tony
Hope you also will test how 4x1GB overclocking does too on RD600 :)
Nice you read my mind, Tony please try with 4x1Gb ram if you can. :DQuote:
Originally Posted by eva2000
So lets see here
1) 511 FSB at least 4 hours orthos stable (bingo13 was this small/large FTT's or blend?)
2) SLI hacked drivers work
3) Faster in 3D than 680i
4) memory clock is true async
5) up to 1.94 Vcore available
6) vdroop of .006 @ load seen thus far
Seems to be winner for most of us so far. Only Phase users with less than 10 multipliers will be FSB limited.
just imagine I have a cascade :DQuote:
Originally Posted by Tony
or that I have THE golden chip which is awsome with SS
500 is not really enough :(
get a 6800 thenQuote:
Originally Posted by metalazzo
Im pushing 866 4-4-4-12 1T at the moment, infact this is where I installed XP on the raptors so its kinda stable ;)
I had a chat with Oskar earlier, he says i should see over 9001T once I get used to the board. He is pushing for 1000 1T 32M pi stable ;)
1T seems the way to go here with the chipset being a little slower than 975 in 2T.
Ive not seen many E6600 doing over 9x511 on ss phase except for 1M maybe. I think most that would be after the last few mhz only for 1M would pick the faster RD680 board anyway (if that one also does 511).
1. Blend, and it made it to 6:53 before giving up the ghost. I need to put a fan on the northbridge I believe. You have to zap the NB with volts to get above 480 or so. ;)Quote:
Originally Posted by The Nemesis
4. 98% true async, a few settings are not available, example, skip from 433 (866) to 435 (870).
bingo13, you wrote you have used 7 images, is that because your RAID got corrupted after locks/reboot as a cause of experiment with OC, OR you see SATA/RAID errors/lockups while running HDD untils (like with current 680i mobos)?
oh, that's it ? :clap:Quote:
Originally Posted by Tony
this mb will rock with 6700 or 6800 that's true
but those cpus are a bit expensive ;)
well, I'll give it a try if I can buy one ;)
Tony you said; I will ask DFI netherlands for stock forcast tomorrow.
Any information about that?
Do we realy have to wait till Feb?
I forgot to ask, but a tech rep from DFI Netherlands did post in this thread quoting new year so lets wait and see.
New Year? :( I'm afraid I can't wait that long.
Orthos is crashing ABIT AW9D-MAX as well but noone want's to bellieve and as many times as I told them, that it might be a program's error and not mobo's error.....They can do everything but when they come with Orthos, it crashes.....:(Quote:
Originally Posted by Tony
Remember; Asus Commando, P965, abit IN9 32X-MAX nF680Quote:
Originally Posted by Trice
Both will launch in January.
Nice start of the new year, but ok I also want a DFI RD600 NOW :fact:
I'm sold! Time to get rid of the Evga 680i :)Quote:
Originally Posted by bingo13
Gettin around 64ns latency in Everest at 900 4-4-4-4-12 1T, how does that compare with i680?
Anyone got any results to compare too?
try good old prime95 v24.14 dual instances instead.. i have to revert to this when testing 4x1GB as orthos for winxp pro sp2 32bit only seems to allow allocating a max 1GB per process even with /3GB switch, while prime95 v24.14 allows full use of 3GB usable by windows apps.Quote:
Originally Posted by hipro5
Hey tony, will those timings in 1T get any tighter? 900mhz 3-3-3 would be a golden bench setting.
In 1t will it beat 975 in super pi?
I AM one of those super pi people.
Timings for 900 1T, 8000VX BTW ;)
Fatty may do 3-3-3 at 900 1T, this newer D9 has no chance i feel.Quote:
Originally Posted by Revv23
Im getting 4010MB/s in memtest86+ at 900 4-4-3-4-12 but its just not quite stable :(
No....Some of them play with cascades or LN2....;)Quote:
Originally Posted by Tony
Though I think that we will need more income from the owners themshelves when they buy it coz your CPU might do 511MHz MAX and mine can do 552MHz max.....So maybe the chipset could handle/work at 552MHz fsb if our CPU could do it......;) Only time will tell though....:)
BTW, have you tried this CPU on a P5B-Deluxe to see how high fsb is reaching?....
What's the word on max FSB improving with new bios revisions?
Ive been without my desktop since October......I have a 628B 660 sitting here as well as 2 GB of RAM, and all I need is a MB!! I just want to run 3.6 and RAID 0!Quote:
Originally Posted by matinez
You coulda been on just about any higher end 965P or i975 and had that.Quote:
Originally Posted by Trice
No the clockgen won't do it.Around 520 is going to be the max.Quote:
Originally Posted by hipro5
I can do 550+ here with a lower multi on other boards.
Quote:
Originally Posted by Trice
I got a 628B E6400, Crucial 10th Ann, 3x 80GB just need a mobo!
I am somewhat confused. Does this ansync. memory ability benefit people with slower DDR2 as well? Say, PC4300 (DDR2)?
Are any stores confirmed to have this on the 15th? or over the weekend?
Im ordering my mobo before christmas, and its either this or an asus p5w-dh deluxe.
Yes it would, because it would enable you to keep your RAM at whatever speed it's most comfortable at, no matter how low that may be.Quote:
Originally Posted by Killnine
Just a quick update -
I am running 3-3-3-6 1T at 866 currently with the E6600 at 7x511. This setting will pass every game benchmark, 3DMark, or PCMark we have run including a loop of 3D06 with Nero Recode running in the background converting a 8GB movie down to 4.7GB.
We still have some online game play to complete but the reason I bring this up is that Orthos and Dual 32M SuperPI will each fail within two minutes of being started. Also, under Orthos and the 3DMarks, my 7x511 (3577MHz) reads as 4579MHz, great for screenshots but not so great for accurate readings. ;)
What is more important then? Application stability or Orthos stability? I vote for application stability as my last Orthos stable (6:53 h/m) setting failed three out of five game benchmarks along with 3D03 of all things before I started over. :confused:
So if your CPU can do about 550MHz fsb on another mobo, then it's safe to say that this mobo can do from 500MHz fsb to 550MHz fsb depending on chipset.....No?......:)Quote:
Originally Posted by Tony
NOT bad at all I would say.....;)
can't wait to put some fatbodies in their.Quote:
Originally Posted by bingo13
Quote:
Originally Posted by hipro5
You need to read again, the board will not do 550fsb as the clockgen will not do it.
You will need to stay around 520 tops and adjust multi to suit.
Yes I understood what you said.......But what if a newer clockgen could do the job?....:)Quote:
Originally Posted by Tony
P.S. GO to sleep....It's TOO late even in England.....:D
I'm at work right now so I can't give a comparison at the same clocks till I get home. but here's what I can get from previous benches. These were done before CPU-Z & Everest could recognize the 680i.Quote:
Originally Posted by Tony
E6400 465FSB 4-4-4-12 2T
EVGA 680i
2GB Kit Crucial 25th Anniversary
http://img.techpowerup.org/061119/ca... 465fsb716.png
E6400 475FSB 4-4-4-12 2T
BFG 680i
2GB Kit Crucial 25th Anniversary
http://img.techpowerup.org/061119/ca...475 X 8848.png
Tony did not wake up until four, he still has a few hours to put in before he can even consider taking a break. :D Just kidding, this board is going to ruin our sleeping patterns for the next week or so.Quote:
Originally Posted by hipro5
You want to see how your rig handles heat under load try intel's TAT @ 100% on both cores. I was surpised to see it was more brutal than Orthos blend.
If we are just considering application stability and gaming I could claim 3.7-3.8. 3.7 will last about 1 hour 40mins under Orthos blend. 3.8 much less :p:
I'll be interested to see how much real preformance this board delivers over 965|975.
Hi bingo, do you think this could be due to some default ram timing too tight? and hopefully when more user get hold on this MB will crack out this problem.Quote:
Originally Posted by bingo13
hey bingo, what ram volts are you at? it would be sweety if exreme voltage took it even further for the benching crowd.
If its not orthos stable than its not orthos stable.Quote:
Originally Posted by bingo13
Orthos is much harder to run than 3dmark/encoding/etc, and its sad that it cannot even run dual 32m, because dual 32m is rather easy to do as a whole.
But doesn't running memory asynchronously mean a slight performance hit with a NB memory controller? Or is this somehow no longer an issue with this system bus arch? Sorry, it's been a while since I've been on the dark side. ;)Quote:
Originally Posted by EnJoY
It seems better than 680 at memory performance but worst than 965..moreover 975X!
The issue is some of my orthos stable settings are not game or application stable, hence the fact that unless you run orthos for a living, then what difference does it make if it is orthos stable or not. :D This has been going on for a long time now and I have serious doubts about the Orthos stable tag being the end all for reporting stability on a system. If I can run Orthos for eight hours but cannot play BF2 online for more than 15 minutes then I would hardly consider my system stable. Just my viewpoint and opinion on the subject. ;)Quote:
Originally Posted by fhpchris
I am at 2.48V at those settings.Quote:
Originally Posted by Revv23
There is no default timing on this board, everything is open for change. :D Actually, I am running the board/memory on its ragged edge at those settings. I would have been surprised if it had passed Orthos or SuperPI but I was surprised that anything else we threw at the board ran perfectly, 26 different games, seven synthetic benches, nine different audio/video applications, a dose of Office 2007, and just for fun watching a little Talladega Nights: The Ballad of Ricky Bobby on a Plextor Blu-ray drive this evening while downloading 8GB of data off the server.Quote:
Originally Posted by argonmtt
By the way, I am running audio, RAID 0, X1950 CrossFire, and have an external hard drive attached to the number 5 USB port and another one on the Firewire port transferring our test scripts in the background. While it failed Orthos at those settings, it has not even blipped while doing application work. I am finding that this chipset performs best with high memory speeds above 1100 with 4-4-4-12 timings better than my 866 3-3-3-6 1T timings in applications but not in SuperPI. ;)
Sounds like a pretty decent run that you're doing over there bingo13.
Keep it coming m8
Interesting, perhaps this board could be a winner on both fronts?Quote:
Originally Posted by bingo13
Wins super pi at 1t, and then goes on to win 3d at 2t?
Quote:
Originally Posted by bingo13
Hi Bingo,
Was the issue on Orthos blend or large and small fft tests?
Trouble with application stable is sometimes providing evidence en masse in a write up. There appears to be no real software the hammers every section of the motherboard simultaneously and provides a report at the end, that's the only reason I use it.
bf2 fails at stock too sometimes..
the other question is that if orthos blend passes at stock, why does it fail at overclock?..
I've had 3d marks pass and games fail too, orthos pass and games fail, you name it. Perhaps this spells the need of someone to step forwards with some new testing software that generates an evidence of pass, and can fulfil a real stern test of Conroe/chipset/ram/gpu all neatly in one package..
regards
Raja
Quote:
Originally Posted by raju
++++++++++++++++
I second this.. :) nice one Raja..