ECS motherboards

  • Thread starter Thread starter Steven Liburd
  • Start date Start date
S

Steven Liburd

I'm looking into upgrading my system, on the cheap. I've seen an ECS
nForce4, Socket 939 board (combo with an Athlon 64 3000+, 256Meg).

1) Is ECS not worth buying?

2) Is 'dual channel' a function of the board, or the RAM, or both?

I'm trying to go cheap, but I at least want to get what I'm paying for...TIA

==steven
 
I'm looking into upgrading my system, on the cheap. I've seen an ECS
nForce4, Socket 939 board (combo with an Athlon 64 3000+, 256Meg).

1) Is ECS not worth buying?

They're not on my short-list, but opinions are like...
2) Is 'dual channel' a function of the board, or the RAM, or both?

The memory controller. In the case of AMD, the memory controller
is on the processor. Dual-channel takes pairs of standard DIMMs. One can
usually use single DIMMs, but won't get the advantages of Dual channel.
I'm trying to go cheap, but I at least want to get what I'm paying
for...TIA

Information is always a good idea, but there is a reason some are much
cheaper than others. I wouldn't skimp on the motherboard. It's too
important. In your case I might go light on memory (256MB is already
light though) with a single stick and add another when you get a little
more money. Be sure to check the MB manual (most are online) to see if
it'll work with a single stick though.
 
I'm looking into upgrading my system, on the cheap. I've seen an ECS
nForce4, Socket 939 board (combo with an Athlon 64 3000+, 256Meg).

1) Is ECS not worth buying?
Well they sure sell a lot of them for being worth nothing. I think they
sell more than most all the other manufacturers combined. i've used
probably 20 or 30 of them over the years. it my experience that the
quality of about all mainboards appear about the same. The difference
between most being add-on features. So if the board has all the features
you want, get it.
2) Is 'dual channel' a function of the board, or the RAM, or both?
Dual channel is built in to the 939/940 cpu's. The board must provide
physical connections and support for it. A ram stick is a ram stick,
nothing more. It doesn't know dual channel from single channel. All
references by marketing to dual channel ram is just BS meaning you get 2
sticks of the same ram. The only advantage being that you get 2 sticks of
the same type that shouldn't have a problem running with the same timings
on seperate channels.
I'm trying to go cheap, but I at least want to get what I'm paying
for...TIA
Don't we all. If you plan on overclocking, then you need to pick a board
that provides for it. If not, just find one that does what you want.
 
I never look at ECS because they made a board with the wrong resistors and
when a site posted how to fix it, they made them take it down.


--
Ed Light

Smiley :-/
MS Smiley :-\

Send spam to the FTC at
(e-mail address removed)
Thanks, robots.
 
I never look at ECS because they made a board with the wrong resistors and
when a site posted how to fix it, they made them take it down.

I'm sure this is something you heard, but in fact it can't be true. Think
about it. They may have asked them to remove it for some reason, but they
sure have no legal means of forcing them to do so.
 
Steven said:
I'm looking into upgrading my system, on the cheap. I've seen an ECS
nForce4, Socket 939 board (combo with an Athlon 64 3000+, 256Meg).

1) Is ECS not worth buying?

I'm a big fan of ECS. I've had this ECS K7S5A motherboard for four
years. It's one of the most reliable I've ever had, even though it was
also one of the cheapest. No overclocking capability in its default
BIOS, but that's probably why it's so reliable. I put the same
motherboard into my brother's machine, and he's been running fine for
3.5 years.
2) Is 'dual channel' a function of the board, or the RAM, or both?

It's a function of the memory controller, which in the case of an Athlon
64 is inside the CPU. On Intel PCs, that controller is inside the chipset.

It also has a little to with the motherboard, as the motherboard has to
have enough traces coming from the RAM slots to do dual channel. But I'd
guess most Socket 939 boards would have the right number of traces as
that's the biggest selling point of Socket 939 over the previous Socket
754.
I'm trying to go cheap, but I at least want to get what I'm paying
for...TIA

I've had no problems with my old ones. But that's no guarantee of
present or future quality.

Yousuf Khan
 
Wes Newell said:
I'm sure this is something you heard, but in fact it can't be true. Think
about it. They may have asked them to remove it for some reason, but they
sure have no legal means of forcing them to do so.

It's true. It was that K7S5a, I think. Resistors under the cpu socket.
memtest86 would give errors until the resistor fix.

They had a thread and the pdf fix tutorial, until they posted that ECS
threatened them, and they backed off like a mouse.


--
Ed Light

Smiley :-/
MS Smiley :-\

Send spam to the FTC at
(e-mail address removed)
Thanks, robots.
 
Ed Light said:
It's true. It was that K7S5a, I think. Resistors under the cpu socket.
memtest86 would give errors until the resistor fix.

They had a thread and the pdf fix tutorial, until they posted that ECS
threatened them, and they backed off like a mouse.

And ECS said they didn't believe in memtest86.


--
Ed Light

Smiley :-/
MS Smiley :-\

Send spam to the FTC at
(e-mail address removed)
Thanks, robots.
 
It's true. It was that K7S5a, I think. Resistors under the cpu socket.
memtest86 would give errors until the resistor fix.

They had a thread and the pdf fix tutorial, until they posted that ECS
threatened them, and they backed off like a mouse.

Well, they may have threatened them, but I found the "fix" in several
places with a simple k7s5a fix search. I still think somebody fed you FUD.
You'll find the fixes here.

http://www.geocities.com/mrathlon2000/
 
Wes Newell said:
Well, they may have threatened them, but I found the "fix" in several
places with a simple k7s5a fix search. I still think somebody fed you FUD.
You'll find the fixes here.

http://www.geocities.com/mrathlon2000/

It is my own personal experience with the web site. So, there is no 3rd
person.


--
Ed Light

Smiley :-/
MS Smiley :-\

Send spam to the FTC at
(e-mail address removed)
Thanks, robots.
 
It is my own personal experience with the web site. So, there is no 3rd
person.

So it was you that caved in to them, or your employer? Looks like others
didn't. Or maybe I misunderstood and you were using their (ECS) website.
If that's the case, then they can do what they want with it.
 
Wes Newell said:
So it was you that caved in to them, or your employer? Looks like others
didn't. Or maybe I misunderstood and you were using their (ECS) website.
If that's the case, then they can do what they want with it.


OK ... starting over ...

I was browsing the net, came upon a website, it had a forum, one thread was
the ECS m/b problem, they worked out a solution and posted a pdf, then ECS
came along and said no no bad things will happen to you you horror, then the
thread and pdf were gone.

Nobody told me about it -- I watched it happen.
--
Ed Light

Smiley :-/
MS Smiley :-\

Send spam to the FTC at
(e-mail address removed)
Thanks, robots.
 
I have about a half dozen ECS motherboards that work well...some of the
time. I just can't seem to synchronize their work schedule with mine.
Most are fine until turned off. Then they won't turn on until they're
stone cold. Once re-started they'll run for weeks.

I'm looking to replace these with reliable hardware.

B
 
I have about a half dozen ECS motherboards that work well...some of the
time. I just can't seem to synchronize their work schedule with mine.
Most are fine until turned off. Then they won't turn on until they're
stone cold. Once re-started they'll run for weeks.
Sounds like weak PSU's. Once started, the load on the PSU is much less
than startup. Of course they heat up while they run and then there
efficiency drops, so if you turn them completely off and try to turn them
back on imadiately they are still hot and don't have enough power to start
up the PC components without overloading.

Properly configured I've never had any more problems with ECS boards as
I've had with any other brand. ECS OEM's boards for many MB companies. As
a matter of fact your XYz brand may be made by ECS.
 
Other motherboard manufacturers subcontract ECS to make motherboards for
them.
Asus gets motherboards from ECS and sells them under the name AsRock.
My PcChips M848ALU is made by ECS (L7S7A2), works fine.
So I would buy an ECS board again.

ECS got a bad rep. from their K7S5A boards which would lose BIOS settings.
The local Fry's Electronics would have stacks and stacks of returned K7S5As.
Their KT-600 seems to be a good basic motherboard.
 
Tod said:
Other motherboard manufacturers subcontract ECS to make motherboards for
them.
Asus gets motherboards from ECS and sells them under the name AsRock.

Maybe, though, that's AsRock's design, not ECS's.
ECS got a bad rep. from their K7S5A boards which would lose BIOS settings.
The local Fry's Electronics would have stacks and stacks of returned
K7S5As.
Their KT-600 seems to be a good basic motherboard.

A web site had a forum thread on the K7S5A and a fix for the memtest86
failures. But ECS threatened it and it had to remove the thread and the fix.
Since then I've avoided ECS.


--
Ed Light

Smiley :-/
MS Smiley :-\

Send spam to the FTC at
(e-mail address removed)
Thanks, robots.
 
A web site had a forum thread on the K7S5A and a fix for the memtest86
failures. But ECS threatened it and it had to remove the thread and the fix.
Since then I've avoided ECS.

Come on Ed, you've got to get off this kick as I've pointed out many
websites that have this fix and others for the K7S5A board. You said you
saw it online, but can you be sure it was actually ECS. Perhaps you were
mistaken, or someone posed as ECS, which any 10 year old could do. Once
again.

Results 1 - 10 of about 23,400 for k7s5a fix - 0.33 sec

And Number 1.

K7S5A M830LR Memtest86 Data corruption problem fix
Memtest86 and Data corruption problem fix. To fix this problem, add a 200
to 220 Ohm resistor in parallel to the 56 Ohm ZP resistor. This makes the
ZP resistance about 46 Ohms and generally solves the data corruption
problem.
 
The 6 ECS MB's that I have are all K7S5As. The PSUs are all from
different manufacturers and I have monitored the voltage because I too
thought that this might be the problem. It wasn't.

As for ECS getting a "bum wrap" I'd suggest that since they sold them
and didn't recall them when there apparently was a noticeable problem
their ethics and technical competence should be questioned. I'd say
they didn't suffer much.

I'll never buy from them, and will be cautious to make sure that I don't
get a MB made by them and branded for someone else.

B
 
Back
Top