Another on the way.

[Jon]

Just confirmed an order on another A7M266-D. Will be here by Tuesday but will have to wait until Friday for everything else. Only going to have dual Duron 1ghZ in this for now, but the upgrade path is much nicer than the ECS dual PIII 1GHz board I sold.

Probably won't get much extra done with it but should shave at least an hour off the dual PIII times. I just hope those chips can hit 1.2Ghz for me.

[Cy]

We have a couple of systems similar to that at my work and they are doing around 5-5.5 per w/U. Should be a nice little cruncher. :)

Cy

[Morpheus]

I might have to add a little fire power myself...

;)

[hallen]

Originally posted by Jon
I just hope those chips can hit 1.2Ghz for me.

Nick from Greece is getting 1.26 from his.

http://forums.overclockers.ws/vb/showthread.php?s=&threadid=58314

[hallen]

Originally posted by Morpheus
I might have to add a little fire power myself...

;)

Try these www.newegg.com

[Jon]

Originally posted by hallen


Nick from Greece is getting 1.26 from his.

http://forums.overclockers.ws/vb/showthread.php?s=&threadid=58314

May be a bit harder on an MPX with only 1.85V to work with though. I may actually try a volt mod on this one though to see how it stands up.

[hallen]

Can you get 1.85 by BIOS alone or are you planning on connecting the L11 bridge?

[hallen]

Because if the BIOS gives you + .10v then connecting the bridge starts you at 1.85v for a total of 1.95v.

Anyway good luck. There are lots of BP6 owners looking for someplace to go.

Harvey

[Jon]

Board only has voltage adjustments for one CPU (weird stuff) so I'll have to close bridges for 1.85V.

There is a new revision of this board apparently out now that allows +.10V on both processors, but it's highly unlikely that the one I'll be getting is one of those...I can only hope.

I'm not too concerned with a huge overclock out of it anyway. I have every intention of upgrading those processors to XPs in the future sometime. Probably drop the ones I have now in there and get some 2000+ for my main one, at some point, then sell off the Durons, or whatever.

At least I have some options now...

[hallen]

Jon- Does the A7M266-D have /5 PCI divisor? I can get 174MHz FSB from an A7M266-single with KingMax PC2700 but /4 divisor limits me to IDE disk and Matrox G200 AGP.

Pair of XP 2000+'s running at 174MHz fsb that would make the room shake.

[Jon]

Nope. Even though the names are almost identical, they are totally different chipsets.

I could run both of my XP 1600s at 160MHz bus for 1.68GHz long enough to run any bench, but was still too unstable for normal use. Has to be the registered ECC or not enough VCore casuong the instabilities. Had no blue screens or lockups at that speed, just random IE crashes and other programs.

I may try running some regular unbuffered in one of them to see if I can narrow it down. Got a couple sticks that do 160MHz easy, I just never have time to tear into them and try.

Found the information I needed on the volt mod for it so I may juice up this new one if I can get some more grabbers for the IC legs.

I couldn't imagine what one of these boards would be like on a 170+MHz bus...:burn:

[hallen]

Despite late start, Epox catches up in server industry

Walson Hu, Taipei; Willie Teng, DigiTimes.com [Thursday 31 January 2002]

Although Epox Computer only entered the server market in the fourth quarter of 2001, it has already launched 1U servers running on the Pentium 4 (P4) and its Advanced Micro Devices (AMD) dual-CPU servers have begun limited shipments as well.

The Taiwan-based company is projected to ship 5,000 P4 servers and few thousand AMD-based servers per month.

Despite its limited time in the industry, Epox is the first to introduce 1U servers running on the P4. Most of the mainstream 1U servers in the market today still run on the PIII. The company has plans to launch servers powered by Intel?s Xeon processor.

Mass production for AMD-based servers is scheduled to begin in early February. Epox separated the AMD models into two series, A and B. The A series include 2U, 3U and tower servers, and the B series 1U models.

For industrial computers, Epox is expected to produce a small amount for a large variety of models in 2002. The company is expected to ship 3,000-4,000 units this year, up from 1,000 units in 2001. Industrial computers currently account for a mere 1.5% of Epox?s revenues, but the number is expected to rise to 20% by year-end.

I wonder what motherboard Epox is using in its AMD servers

http://www.digitimes.com/NewsShow/NewsSearch.asp?DocID=D8D00509A339414D48256B51004B9 927&query=AMD

[Jon]

I wonder what motherboard Epox is using in its AMD servers

It's the M762A. Scroll down just a little here (http://www.epox.com/html/english/default.htm). Looks OK, but I don't know about server quality. Same as all the other MPX boards coming out which I don't consider heavy server quality...yet.

[hallen]

Lots of interesting stuff on Epox front page. Thanks. Not much on the M762A
though.

I have another question. Your farm page says you get 12+ wu's/day from the A7M266-D with dual XP 1600's @ 150MHz fsb.

I get about 7.5 wu's/day from a 1600 so that seems like pretty efficient use of the second processor. That would mean that your're getting times like 1:50 to 2:10 per wu.

Those times are pretty steady day in and day out?

Also does using two processes improve or slow things down?

Thanks

Harvey

[Jon]

I'm only averaging 3h45m per WU. I get at least 6 WU per day from each CPU so it's actually a little less efficient than yours is on a single board. If you're getting 7.5 a day from yours then two would get 15...sharing the same bus and such does hinder performance a bit. Does the same thing on any dual platform. Lost about an hour and half per WU on dual PIII 1GHz boards.

There were a couple reviews on the EPoX board somewhere. It will be the first MPX board released with functional onboard USB. Didn't do any better/worse than the others that are out though.