Slashdot Mirror
First Intel Quad Core Ready Desktop Mobo Spotted
MojoDog writes "Today Universal Abit launched their AW9D and
AW9D-MAX motherboards based on the Intel 975X chipset. There has been much
anticipation in the industry for this series and as far as looks go, these
boards are built to please. One interesting bullet point in the spec list
is that these boards are "Quad Core Ready", in line with a possible year-end
release of
Intel's Quad-Core Kentsfield CPU perhaps? Time will tell!"
Will that make the Mac Pro any cheaper? Or maybe a quad-core iMac a possibility?
I have a AB9 Pro Intel 965-based board here with a Core 2 Duo E6400, and I can't get it to boot half the time. I get an error code 8.7. on the motherboard's LCD, which means "Check CPU Core Voltage". When it does boot, I occasionally get an error or "Device Verify Failed" from the AHCI BIOS while identifying my hard drives.
The system is impressively fast when it actually boots and works, but those two issues make the motherboard very difficult to actually use.
What is "Universal Abit"? I always thought the company was just called "abit". But my ignorance is boundless.
How can a motherboard have all this stuff and leave out ECC? I would never buy a motherboard without ECC. Don't people want their machines to stay up more than a week at a time???
Your god may be dead, but mine aren't!
"All you have to do is be fragile and grateful. So stay the underdog." Chuck Palahniuk, Choke
Not terribly suprising for a new board, considering that existing boards can already support early kentsfield chips. Including a kentsfield benchmark run on a MSI 975X Platinum 2.b
Of course some people are always going to want to have the fastest game machine on the block. But seriously, it's amazing the kind of performance you can get these days with cheap, low-end hardware. Yesterday I built a machine for $300 with a 3 GHz P4 and 1 Gb of ram. (I reused a hard disk, so that cut the price a little.) Sure I could have built a dual-core system, but I would have ended up with a machine that cost many times more, used tons of power, and had almost identical performance except when I had two cpu-intensive processes running at once (i.e., almost never).
Find free books.
I'm not ready to state that GA-965P-DQ6 is the "fist", but it's the first one I found out about (months ago). I've already got this MB running with a Core 2 Duo E6600, and it specifically states on the manufacturer "Ready for next generation Quad Core processor".
P roducts_Overview.aspx?ProductID=2295
http://www.gigabyte-usa.com/Products/Motherboard/
I vote to change the name of the article "Intel Quad Core Ready Desktop Mobo Spotted" to "Another Intel Quad Core Ready Desktop Mobo Spotted".
I'd guess that there are at least a few Apple employees who have been looking at quad-core-ready boards for some time now :)
The quad-core Kentsfield is a subsitute for the previous anounced 3.20GHz Core 2 Extreme, which has been cancelled:1 24626.html
http://www.xbitlabs.com/news/cpu/display/20060817
A new generation of processing power is coming
to us in less than 6 months!
2 physical CPUs on a Mainboard with each
2 Woodcrest units, 4MB 2nd each with each
2 cores
8 cores!
I'm looking forward to it. Put 16GB RAM into it and you'll
have a perfect setup for huge Websphere/Weblogic cluster
tests on a _single_ workstation!
See: http://pics.computerbase.de/1/3/0/0/4/2.png
Coooool!
I'm still going to wait for the eight-core boards, until then I will cherish my 486.
Intel has implied that any board that supports this wave of Core 2 chips should have no trouble with Kentsfield, and IIRC all 965-chipset boards are confirmed as such. And most of those boards don't cost near what this one does.
This is a complete and total non-story.
Never mistake "can" for "should".
For those who use their computers for finding aliens, cures for diseases or other distributed computing projects, quad core machines would be great.
The amount of potential distributed computing power available has to be staggering, and its growing every day.
Probably both. AMD released there press announcement on Tuesday. Intel, are you slacking off or are you not as fast as AMD?
The Rapture is NOT an exit strategy.
Now way! That's too many calories.
In queuing theory, and in practice, it is well known that in a multiple server environment, the shortest time to service occurs when there is a single queue facing multiple servers. The first task in the ready queue goes to the next available server. Service is from task initiation to task completion.
With this above configuration, the average queuing time, and hence service time is much less than if there is a dedicated queue per server.
From what I have seen, with multiple core systems, there appears to be a dispatcher per server, so that queues are the type that you find in grocery store check out counters (a queue per server). Throughput time is longer than if there was one server dedicated to dispatching and the others dedicated to running tasks.
The other thing that comes out of all the research and experience, is that 2 servers that execute twice as fast as the 4 servers provide higher throughput.
So, why the big fuss about quad core, unless it is to save real-estate or electrical connections. Give me dual core that executes twice as fast and I will be happier.
I actually think that a three core system, (a dispatcher cpu, and two server cpus) would be just fine. Now, to confirm that all operating systems use a single dispatcher and multiple server configuration I count on the operating system developers to check things out.
FYI. My background is that I have more than 40 years in Computer Capacity Planning, System Tuning and system development. I am a mathematician (applied) who uses this talent to my max.
Leslie Montreal
Leslie Satenstein Montreal Quebec Canada
I like, "Slashdot Recommended Intel Quad Core Ready Desktop Mobos from Abit."
For non critical application systems, (home computers, or systems with small databases, you can do without ECC memory. Just as you can do without car insurance.
Suppose you have a terrabyte database and you have not used ECC memory. Suppose that a single bit error got into the database undetected. And some months later, the records or pointers or tree of data with the corrupted bit is corrupted. What would be the cost of repair against the cost of having ECC memory installed?
My understanding of ECC Memory operation.
Memory addresses are divided into words. The simplist design of ECC memory takes a check sum of the bits of a word of memory and of itself. Thus, for example, a 32 bit word would need 5 bits to hold a check sum, but then we have the ecc memory that is needed to store the checksum and of course, its own check sum. That adds at least a few more bits. To check 64 bit words only takes another bit, over using a 32 bit word.
The theory is, that for any single bit error, the error would be corrected on the fly and would be transparent to the application. In more sophisticated systems, the same holds true, but the error would also be reported as a soft error. A diagnostic processor in the system would analyse the ECC errors and distinguish between random soft and persistent hard failures.
ECC hardware detection would detect errors with more than a single bit would halt the system.
When ECC memory systems are well designed, you can pull out a card with faulty memory and replace the card with a repaired one. System design would have the card only support a single bit in a word, distributed over many the words of memory. After replacement, the access of memory would result in the rewrite of memory in the replacement card, resulting in generating the correct bit setting.
Our lowly PCs do not have that sophistication, but some mainframe systems do.
Leslie
If I am wrong, please let me know.
Leslie Satenstein Montreal Quebec Canada