Welcome to PCForumz.com!
FAQFAQ      ProfileProfile    Private MessagesPrivate Messages   Log inLog in

Questions about DDR RAM

 
Goto page Previous  1, 2, 3, 4, 5
   Hardware Problem Solving Community! (Home) -> Chips RSS
Next:  Canon Elan lens compatiable with Canon Digitial R..  
Author Message
~misfit~

External


Since: Oct 03, 2007
Posts: 12



(Msg. 16) Posted: Wed Oct 03, 2007 9:53 pm
Post subject: Re: Questions about DDR RAM [Login to view extended thread Info.]
Archived from groups: alt>comp>hardware>homebuilt, others (more info?)

Somewhere on the interweb "Igor" typed:
> On Tue, 02 Oct 2007 20:32:30 -0400, CBFalconer <cbfalconer.DeleteThis@yahoo.com>
> wrote:
>
> <snip>
>
> > But, if the MB can handle it, ensure you get ECC capable memory.
> >
>
> My motherboard's manual doesn't say anything about supporting ECC, so
> I'll assume that it doesn't.
>
> However, I plan on putting together a PC from scratch in the near
> future, so perhaps ECC support is something I should be looking for
> in a motherboard.
>
> In a nutshell, why is ECC capable memory more desirable?

http://en.wikipedia.org/wiki/ECC_memory#Error-correcting_memory

Not the best description but it will suffice.

In a nutshell, to quote from the Wiki page:

"ECC memory provides greater data accuracy and system uptime by protecting
against soft errors in computer memory."

However, it comes at a price. It depends on what you're using your PC for
really. I have two PCs with ECC RAM and the rest without. IMO for
non-mission-critical machines its' not really needed.
--
TTFN

Shaun.

 >> Stay informed about: Questions about DDR RAM 
Back to top
Login to vote
Yousuf Khan

External


Since: Dec 17, 2005
Posts: 143



(Msg. 17) Posted: Thu Oct 04, 2007 9:59 pm
Post subject: Re: Questions about DDR RAM [Login to view extended thread Info.]
Archived from groups: per prev. post (more info?)

Igor wrote:
> I'm planning to add some memory to a PC. The motherboard supports "DDR
> up to 333 MHz (enhanced) memory bus."
>
> 333 MHz (PC2700) or slower DDR SDRAM isn't easy to find on today's
> market, and when you do find it, it's considerably more expensive than
> the DDR which runs at 400 MHz (PC3200). What would happen if I bought
> 400 MHz DDR RAM and installed it in a motherboard which only supports
> speeds up to 333 MHz (PC2700)? Would it work? If it did work, would it
> work as well as the slower PC2700 RAM?
>
> I've gotten conflicting answers to this question. Two computer stores
> told me that it wouldn't work. Another store and a book I consulted said
> it would.

I'll bet the boys in the first two shops were confused between DDR vs.
DDR2 DRAM. The type of memory you're getting is just DDR memory, and
therefore any memory that conforms to the DDR standard will work on your
mobo. Even slower ones, but they'll just run slower on your system, of
course.

Currently, the most popular type of memory is DDR2, so the older DDR
memories are being phased out, and likely the only stock that is left is
the fastest versions of DDR. So that's why you can't find DDR-333/PC2700
memory anymore, because now they only stock the fastest DDR-400/PC3200
memories. On the other hand, if you were interested in DDR2 memory, then
you'll have all kinds of choices such as DDR2-533/PC4200,
DDR2-667/PC5200, or DDR2-800/PC6400.

> I'd also like some opinions on which brand of RAM I should get. Kingston
> and Samsung are ubiquitous where I live, so the choice comes down to one
> of those. Both brands are about the same price, with the Samsung being
> marginally cheaper.

Brand-name doesn't matter, just make sure you buy two of exactly the
same brand at the exact same time, if you're going to be matching it up
for dual-channel operation, otherwise that'll show up as single-channel.

Dual-channel vs. single-channel is another way of increasing the speed
of Dram, with dual-channel being faster than single-channel, obviously.

> Finally, would a 128 MB stick of 266 MHz DDR RAM be able to run in the
> same computer as, say, a 512 MB stick of 400 MHz DDR RAM? If so, would
> using both at the same time degrade the performance of the faster RAM?

Yes, the faster RAM would be slowed down to the speed of the slower RAM.
So it's best to completely remove the old RAM, if you want to gain the
performance advantage of the newer RAM.

As for mail-ordering within Canada, try this site:

www.pricegrabber.ca

Personally, I prefer to do my business with bricks'n'mortar stores
rather than online.

Yousuf Khan

 >> Stay informed about: Questions about DDR RAM 
Back to top
Login to vote
Andrew Smallshaw

External


Since: Apr 03, 2005
Posts: 42



(Msg. 18) Posted: Fri Oct 05, 2007 11:45 am
Post subject: Re: Questions about DDR RAM [Login to view extended thread Info.]
Archived from groups: per prev. post (more info?)

On 2007-10-03 kony wrote:
> On Wed, 03 Oct 2007 16:30:41 -0000, Andrew
> <andrews.RemoveThis@sdf.lonestar.org> wrote:
>
> >Memory is fairly reliable provided you aren't recklessly overclocking
> >things so I'd consider ECC inappropriate on all but the most mission-
> >critical systems. Unless you are talking about a machine already
> >fitted out with a UPS, hot-swappable RAID, redundant power supplies
> >and preferably an secure, climate controlled machine room to put it
> >all in there are more important risk factors to consider.
>
> None of those risk factors change whether the results of
> data calculations are important.
>
> I would say ECC is appropriate for any even slightly
> critical application, not just "the most mission critical".
> You spend a few bucks more for better PSU, case, processor,
> etc, why not the memory? Your data links usually have error
> correction, as does a hard drive internally, as does the
> processor cache, but main memory is a weak link in data
> integrity.

Well, some of these factors _can_ affect calculations in the same
manner as bad memory - e.g. anything can happen with a brief power
outage that a regular PSU can _just_about_ bridge, including
incorrect calculations in the CPU, bad data on the bus lines or
whatever. However, I'm not questioning the fact that ECC is
extremely useful and yes, and aid to reliability. What I am
questioning is whether it should be regarded as a priority.

How often do you see bad hard drives, or blown PSUs, or have a
machine cut out in a power cut? These are a computer's weak points
so it makes sense to protect them. It's a little more difficult
to diagnose memory errors in a non-ECC PC because they would usually
appear to be much-more-likely software problems.

However, look at a machine equipped with ECC memory. How often do
you see memory errors logged? Approximately never. Cast your mind
back ten or twelve years when even commodity machines had parity
memory. How often did you see the BIOS message saying it had frozen
the system because of a parity error? I only ever saw it once.
While my memory is vague ISTR that I discounted it at the time as
it was from a system with other hardware problems. Ask yourself
why parity was removed from memory. Yes, it did save money but
why did the computer industry en masse decide that it could safely
be dispensed with?

In short ECC will improve a system's reliability, but not by much.
You are far better off spending the money elsewhere, looking at
more likely causes of failiure, and considering backup, redundancy
etc first. Only when those have been covered is it worth worrying
over whether your machine has ECC memory or not.

--
Andrew Smallshaw
andrews.RemoveThis@sdf.lonestar.org
 >> Stay informed about: Questions about DDR RAM 
Back to top
Login to vote
kony

External


Since: Jan 03, 2004
Posts: 6148



(Msg. 19) Posted: Sat Oct 06, 2007 2:47 am
Post subject: Re: Questions about DDR RAM [Login to view extended thread Info.]
Archived from groups: per prev. post (more info?)

On Fri, 5 Oct 2007 15:17:57 +0000 (UTC), Andrew Smallshaw
<andrews.TakeThisOut@sdf.lonestar.org> wrote:

> Yes, it did save money but
>why did the computer industry en masse decide that it could safely
>be dispensed with?
>

Because they don't particularly care if a customer's
calculations/etc end up wrong if it's not guaranteed for
some critical use. The industry didn't actually abandon it
for critical uses.
 >> Stay informed about: Questions about DDR RAM 
Back to top
Login to vote
krw

External


Since: May 08, 2006
Posts: 56



(Msg. 20) Posted: Sat Oct 06, 2007 6:03 pm
Post subject: Re: Questions about DDR RAM [Login to view extended thread Info.]
Archived from groups: per prev. post (more info?)

In article <slrnfgclcb.1aa.andrews.RemoveThis@sdf.lonestar.org>,
andrews.RemoveThis@sdf.lonestar.org says...
> On 2007-10-03 kony wrote:
> > On Wed, 03 Oct 2007 16:30:41 -0000, Andrew
> > <andrews.RemoveThis@sdf.lonestar.org> wrote:
> >
> > >Memory is fairly reliable provided you aren't recklessly overclocking
> > >things so I'd consider ECC inappropriate on all but the most mission-
> > >critical systems. Unless you are talking about a machine already
> > >fitted out with a UPS, hot-swappable RAID, redundant power supplies
> > >and preferably an secure, climate controlled machine room to put it
> > >all in there are more important risk factors to consider.
> >
> > None of those risk factors change whether the results of
> > data calculations are important.
> >
> > I would say ECC is appropriate for any even slightly
> > critical application, not just "the most mission critical".
> > You spend a few bucks more for better PSU, case, processor,
> > etc, why not the memory? Your data links usually have error
> > correction, as does a hard drive internally, as does the
> > processor cache, but main memory is a weak link in data
> > integrity.
>
> Well, some of these factors _can_ affect calculations in the same
> manner as bad memory - e.g. anything can happen with a brief power
> outage that a regular PSU can _just_about_ bridge, including
> incorrect calculations in the CPU, bad data on the bus lines or
> whatever. However, I'm not questioning the fact that ECC is
> extremely useful and yes, and aid to reliability. What I am
> questioning is whether it should be regarded as a priority.

Priority? NO, but for an 11% uplift over *CHEAP* DRAM, it's
surprising it's not standard again. I have ECC memory on my
"desktop" (at a cost *way* over 111%) and would have bought it for my
laptop, if it were available. Bit-rot is a PITA.

> How often do you see bad hard drives, or blown PSUs, or have a
> machine cut out in a power cut? These are a computer's weak points
> so it makes sense to protect them. It's a little more difficult
> to diagnose memory errors in a non-ECC PC because they would usually
> appear to be much-more-likely software problems.

Right. Just because you can blame it on M$ doesn't mean it is. It
is *nice* to know when a component is failing. As you point out,
it's extremely difficult to diagnose memory problems. ECC/Parity
would make it far simpler.

> However, look at a machine equipped with ECC memory. How often do
> you see memory errors logged? Approximately never. Cast your mind
> back ten or twelve years when even commodity machines had parity
> memory. How often did you see the BIOS message saying it had frozen
> the system because of a parity error? I only ever saw it once.
> While my memory is vague ISTR that I discounted it at the time as
> it was from a system with other hardware problems. Ask yourself
> why parity was removed from memory. Yes, it did save money but
> why did the computer industry en masse decide that it could safely
> be dispensed with?

I see them on my "desktop" system. I've come to the conclusion that
they're ghosts cause by a BIOS problem, but they're there.

> In short ECC will improve a system's reliability, but not by much.
> You are far better off spending the money elsewhere, looking at
> more likely causes of failiure, and considering backup, redundancy
> etc first. Only when those have been covered is it worth worrying
> over whether your machine has ECC memory or not.

It's a cheap thing to do to catch a difficult problem to diagnose.
If you've ever suffered a severe case of bit-rot I'm sure you would
be a little more sensitive to sources of errors.

--
Keith
 >> Stay informed about: Questions about DDR RAM 
Back to top
Login to vote
Del Cecchi

External


Since: Nov 18, 2005
Posts: 33



(Msg. 21) Posted: Sun Oct 07, 2007 9:54 pm
Post subject: Re: Questions about DDR RAM [Login to view extended thread Info.]
Archived from groups: per prev. post (more info?)

"kony" <spam.TakeThisOut@spam.com> wrote in message
news:npbeg39tfgj8aon14g1sc9ck6f3t9318nl@4ax.com...
> On Fri, 5 Oct 2007 15:17:57 +0000 (UTC), Andrew Smallshaw
> <andrews.TakeThisOut@sdf.lonestar.org> wrote:
>
>> Yes, it did save money but
>>why did the computer industry en masse decide that it could safely
>>be dispensed with?
>>
>
> Because they don't particularly care if a customer's
> calculations/etc end up wrong if it's not guaranteed for
> some critical use. The industry didn't actually abandon it
> for critical uses.

In fact, for critical usage some companies have gone far beyond normal
SEC/DED error correction in servers.
I suppose I ought to trim the crossposting...
 >> Stay informed about: Questions about DDR RAM 
Back to top
Login to vote
daytripper

External


Since: Nov 18, 2003
Posts: 523



(Msg. 22) Posted: Sun Oct 07, 2007 11:20 pm
Post subject: Re: Questions about DDR RAM [Login to view extended thread Info.]
Archived from groups: per prev. post (more info?)

On Sun, 7 Oct 2007 21:54:21 -0500, "Del Cecchi"
<delcecchiofthenorth DeleteThis @gmail.com> wrote:

>
>"kony" <spam DeleteThis @spam.com> wrote in message
>news:npbeg39tfgj8aon14g1sc9ck6f3t9318nl@4ax.com...
>> On Fri, 5 Oct 2007 15:17:57 +0000 (UTC), Andrew Smallshaw
>> <andrews DeleteThis @sdf.lonestar.org> wrote:
>>
>>> Yes, it did save money but why did the computer industry en masse decide
>>> that it could safely be dispensed with?
>>
>> Because they don't particularly care if a customer's
>> calculations/etc end up wrong if it's not guaranteed for
>> some critical use. The industry didn't actually abandon it
>> for critical uses.
>
>In fact, for critical usage some companies have gone far beyond normal
>SEC/DED error correction in servers.

If you are referring to ECC schemes and not memory mirroring, I'd be
interested in examples that *don't* simply use n Hamming codewords to turn
n-bit-wide full chip failures into correctable events - which really wasn't
that remarkable, revolutionary or heroic when it was first employed - about 20
years ago...

>I suppose I ought to trim the crossposting...

And what fun would that be?

/daytripper Wink
 >> Stay informed about: Questions about DDR RAM 
Back to top
Login to vote
Frank McCoy

External


Since: Feb 22, 2007
Posts: 135



(Msg. 23) Posted: Sun Oct 07, 2007 11:20 pm
Post subject: Re: Questions about DDR RAM [Login to view extended thread Info.]
Archived from groups: per prev. post (more info?)

In alt.comp.hardware.pc-homebuilt daytripper
<day_trippr RemoveThis @REMOVEyahoo.com> wrote:

>On Sun, 7 Oct 2007 21:54:21 -0500, "Del Cecchi"
><delcecchiofthenorth RemoveThis @gmail.com> wrote:
>
>>
>>"kony" <spam RemoveThis @spam.com> wrote in message
>>news:npbeg39tfgj8aon14g1sc9ck6f3t9318nl@4ax.com...
>>> On Fri, 5 Oct 2007 15:17:57 +0000 (UTC), Andrew Smallshaw
>>> <andrews RemoveThis @sdf.lonestar.org> wrote:
>>>
>>>> Yes, it did save money but why did the computer industry en masse decide
>>>> that it could safely be dispensed with?
>>>
>>> Because they don't particularly care if a customer's
>>> calculations/etc end up wrong if it's not guaranteed for
>>> some critical use. The industry didn't actually abandon it
>>> for critical uses.
>>
>>In fact, for critical usage some companies have gone far beyond normal
>>SEC/DED error correction in servers.
>
>If you are referring to ECC schemes and not memory mirroring, I'd be
>interested in examples that *don't* simply use n Hamming codewords to turn
>n-bit-wide full chip failures into correctable events - which really wasn't
>that remarkable, revolutionary or heroic when it was first employed - about 20
>years ago...
>
It's just a hamming-code.
However, about 99% of memory installed on PCs is *not* EEC or even
parity enabled. They all *should* be.

>>I suppose I ought to trim the crossposting...
>
>And what fun would that be?
>
>/daytripper Wink

--
_____
/ ' / ™
,-/-, __ __. ____ /_
(_/ / (_(_/|_/ / <_/ <_
 >> Stay informed about: Questions about DDR RAM 
Back to top
Login to vote
Franc Zabkar

External


Since: Sep 10, 2005
Posts: 125



(Msg. 24) Posted: Mon Oct 08, 2007 7:50 am
Post subject: Re: Questions about DDR RAM [Login to view extended thread Info.]
Archived from groups: per prev. post (more info?)

On Wed, 03 Oct 2007 16:30:41 -0000, Andrew <andrews.TakeThisOut@sdf.lonestar.org>
put finger to keyboard and composed:

>On 3 Oct, 08:55, Igor <donotre....TakeThisOut@nospam.com> wrote:
>> However, I plan on putting together a PC from scratch in the near future,
>> so perhaps ECC support is something I should be looking for in a
>> motherboard.
>>
>> In a nutshell, why is ECC capable memory more desirable?
>
>It has additional bits used to store checksum information so that
>errors can be detected and simple ones corrected. In short, it's a
>reliability thing. Without knowing much about your set up, it's
>impossible to be sure but at three or four times the cost of normal
>RAM I doubt it will be a cost-effective way of improving
>reliability.

ECC just adds one extra RAM, ie 9 chips versus 8, or 72 bits versus
64. Even allowing for economies of scale, I can't see why you would
expect a 3x or 4x price difference.

Kingston 1GB 333MHz DDR ECC Registered CL2.5 DIMM Dual Rank, x8
(US$70):
http://www.ec.kingston.com/ecom/configurator_new/PartsInfo.asp?root=&L...Back=&k

Kingston 1GB 333MHz DDR Non-ECC CL2.5 DIMM (US$54):
http://www.ec.kingston.com/ecom/configurator_new/PartsInfo.asp?root=&L...Back=&k

> If you want to do that on most systems, you're better
>off spending the money on things like fitting a UPS and RAID storage -
>these cover much less reliable elements of the system.
>
>Memory is fairly reliable provided you aren't recklessly overclocking
>things so I'd consider ECC inappropriate on all but the most mission-
>critical systems. Unless you are talking about a machine already
>fitted out with a UPS, hot-swappable RAID, redundant power supplies
>and preferably an secure, climate controlled machine room to put it
>all in there are more important risk factors to consider.

- Franc Zabkar
--
Please remove one 'i' from my address when replying by email.
 >> Stay informed about: Questions about DDR RAM 
Back to top
Login to vote
Del Cecchi

External


Since: Nov 18, 2005
Posts: 33



(Msg. 25) Posted: Mon Oct 08, 2007 9:11 am
Post subject: Re: Questions about DDR RAM [Login to view extended thread Info.]
Archived from groups: per prev. post (more info?)

"daytripper" <day_trippr.TakeThisOut@REMOVEyahoo.com> wrote in message
news:vt7jg315a2gmrs8k1294fnpqmncfia41ec@4ax.com...
> On Sun, 7 Oct 2007 21:54:21 -0500, "Del Cecchi"
> <delcecchiofthenorth.TakeThisOut@gmail.com> wrote:
>
>>
>>"kony" <spam.TakeThisOut@spam.com> wrote in message
>>news:npbeg39tfgj8aon14g1sc9ck6f3t9318nl@4ax.com...
>>> On Fri, 5 Oct 2007 15:17:57 +0000 (UTC), Andrew Smallshaw
>>> <andrews.TakeThisOut@sdf.lonestar.org> wrote:
>>>
>>>> Yes, it did save money but why did the computer industry en masse
>>>> decide
>>>> that it could safely be dispensed with?
>>>
>>> Because they don't particularly care if a customer's
>>> calculations/etc end up wrong if it's not guaranteed for
>>> some critical use. The industry didn't actually abandon it
>>> for critical uses.
>>
>>In fact, for critical usage some companies have gone far beyond normal
>>SEC/DED error correction in servers.
>
> If you are referring to ECC schemes and not memory mirroring, I'd be
> interested in examples that *don't* simply use n Hamming codewords to
> turn
> n-bit-wide full chip failures into correctable events - which really
> wasn't
> that remarkable, revolutionary or heroic when it was first employed -
> about 20
> years ago...
>
>>I suppose I ought to trim the crossposting...
>
> And what fun would that be?
>
> /daytripper Wink

In some sense it is a hamming code or along those lines. but it includes
redundant bit steering, correction of double errors when one is hard and
the other soft, package codes, scrubbing, etc. The whole quote was "far
beyond normal SEC/DED...." eg 64/72 type codes that do single bit
correction and double bit detection.

del
 >> Stay informed about: Questions about DDR RAM 
Back to top
Login to vote
Robert Redelmeier

External


Since: Sep 09, 2004
Posts: 226



(Msg. 26) Posted: Mon Oct 08, 2007 3:45 pm
Post subject: Re: Questions about DDR RAM [Login to view extended thread Info.]
Archived from groups: per prev. post (more info?)

In comp.sys.ibm.pc.hardware.chips Frank McCoy <mccoyf DeleteThis @millcomm.com> wrote in part:
> However, about 99% of memory installed on PCs is *not*
> EEC or even parity enabled. They all *should* be.

Oh, pray tell, why? Do you believe you know the PC
business better than Intel, AMD, Dell, HP, ... who have
decided to manufacture chipsets and computers without ECC?

Do you believe ~50 US$/box is better spent on ECC than on
improved capacitors, mobo layers, cabling, cooling or shielding?

There are always many improvements possible. The key is to
choose the best ones. Not fixate like a kid in a candy store.

-- Robert
 >> Stay informed about: Questions about DDR RAM 
Back to top
Login to vote
kony

External


Since: Jan 03, 2004
Posts: 6148



(Msg. 27) Posted: Mon Oct 08, 2007 4:41 pm
Post subject: Re: Questions about DDR RAM [Login to view extended thread Info.]
Archived from groups: per prev. post (more info?)

On Mon, 08 Oct 2007 18:48:57 GMT, Robert Redelmeier
<redelm.TakeThisOut@ev1.net.invalid> wrote:

>In comp.sys.ibm.pc.hardware.chips Frank McCoy <mccoyf.TakeThisOut@millcomm.com> wrote in part:
>> However, about 99% of memory installed on PCs is *not*
>> EEC or even parity enabled. They all *should* be.
>
>Oh, pray tell, why? Do you believe you know the PC
>business better than Intel, AMD, Dell, HP, ... who have
>decided to manufacture chipsets and computers without ECC?

They sell ECC equipped systems too. Would you believe you
knew better if you chose standardized parts instead of
proprietary motherboards, cases and PSU? Most agree on
that. The idea of blindly following an OEM is contrary to
our goals.
 >> Stay informed about: Questions about DDR RAM 
Back to top
Login to vote
krw

External


Since: May 08, 2006
Posts: 56



(Msg. 28) Posted: Mon Oct 08, 2007 11:06 pm
Post subject: Re: Questions about DDR RAM [Login to view extended thread Info.]
Archived from groups: per prev. post (more info?)

In article <vt7jg315a2gmrs8k1294fnpqmncfia41ec.DeleteThis@4ax.com>,
day_trippr.DeleteThis@REMOVEyahoo.com says...
> On Sun, 7 Oct 2007 21:54:21 -0500, "Del Cecchi"
> <delcecchiofthenorth.DeleteThis@gmail.com> wrote:
>
> >
> >"kony" <spam.DeleteThis@spam.com> wrote in message
> >news:npbeg39tfgj8aon14g1sc9ck6f3t9318nl@4ax.com...
> >> On Fri, 5 Oct 2007 15:17:57 +0000 (UTC), Andrew Smallshaw
> >> <andrews.DeleteThis@sdf.lonestar.org> wrote:
> >>
> >>> Yes, it did save money but why did the computer industry en masse decide
> >>> that it could safely be dispensed with?
> >>
> >> Because they don't particularly care if a customer's
> >> calculations/etc end up wrong if it's not guaranteed for
> >> some critical use. The industry didn't actually abandon it
> >> for critical uses.
> >
> >In fact, for critical usage some companies have gone far beyond normal
> >SEC/DED error correction in servers.
>
> If you are referring to ECC schemes and not memory mirroring, I'd be
> interested in examples that *don't* simply use n Hamming codewords to turn
> n-bit-wide full chip failures into correctable events - which really wasn't
> that remarkable, revolutionary or heroic when it was first employed - about 20
> years ago...

I think you'll find it's been longer than 20 years. More likely
double that.

> >I suppose I ought to trim the crossposting...
>
> And what fun would that be?

Beat me to it.

--
Keith
 >> Stay informed about: Questions about DDR RAM 
Back to top
Login to vote
daytripper

External


Since: Nov 18, 2003
Posts: 523



(Msg. 29) Posted: Mon Oct 08, 2007 11:26 pm
Post subject: Re: Questions about DDR RAM [Login to view extended thread Info.]
Archived from groups: per prev. post (more info?)

On Mon, 8 Oct 2007 23:06:07 -0400, krw <krw RemoveThis @att.bizzzz> wrote:

>In article <vt7jg315a2gmrs8k1294fnpqmncfia41ec RemoveThis @4ax.com>,
>day_trippr@REMOVEyahoo.com says...
>> On Sun, 7 Oct 2007 21:54:21 -0500, "Del Cecchi"
>> <delcecchiofthenorth RemoveThis @gmail.com> wrote:
>>
>> >
>> >"kony" <spam RemoveThis @spam.com> wrote in message
>> >news:npbeg39tfgj8aon14g1sc9ck6f3t9318nl@4ax.com...
>> >> On Fri, 5 Oct 2007 15:17:57 +0000 (UTC), Andrew Smallshaw
>> >> <andrews RemoveThis @sdf.lonestar.org> wrote:
>> >>
>> >>> Yes, it did save money but why did the computer industry en masse decide
>> >>> that it could safely be dispensed with?
>> >>
>> >> Because they don't particularly care if a customer's
>> >> calculations/etc end up wrong if it's not guaranteed for
>> >> some critical use. The industry didn't actually abandon it
>> >> for critical uses.
>> >
>> >In fact, for critical usage some companies have gone far beyond normal
>> >SEC/DED error correction in servers.
>>
>> If you are referring to ECC schemes and not memory mirroring, I'd be
>> interested in examples that *don't* simply use n Hamming codewords to turn
>> n-bit-wide full chip failures into correctable events - which really wasn't
>> that remarkable, revolutionary or heroic when it was first employed - about 20
>> years ago...
>
>I think you'll find it's been longer than 20 years. More likely
>double that.

I'll let you defend that statement with a cite Smile I'm sticking with the
timeframe being the very early '80's, when Digital started shipping systems
using x4 drams in volume and needed to survive a full chip failure...

>> >I suppose I ought to trim the crossposting...
>>
>> And what fun would that be?
>
>Beat me to it.

Bad habit of mine Wink

/daytripper (meanwhile, I'm gonna go dangle flies in front of steelhead Smile
 >> Stay informed about: Questions about DDR RAM 
Back to top
Login to vote
Frank McCoy

External


Since: Feb 22, 2007
Posts: 135



(Msg. 30) Posted: Mon Oct 08, 2007 11:26 pm
Post subject: Re: Questions about DDR RAM [Login to view extended thread Info.]
Archived from groups: per prev. post (more info?)

In alt.comp.hardware.pc-homebuilt daytripper
<day_trippr.DeleteThis@REMOVEyahoo.com> wrote:

>On Mon, 8 Oct 2007 23:06:07 -0400, krw <krw.DeleteThis@att.bizzzz> wrote:
>
>>In article <vt7jg315a2gmrs8k1294fnpqmncfia41ec.DeleteThis@4ax.com>,
>>day_trippr@REMOVEyahoo.com says...
>>> On Sun, 7 Oct 2007 21:54:21 -0500, "Del Cecchi"
>>> <delcecchiofthenorth.DeleteThis@gmail.com> wrote:
>>>
>>> >
>>> >"kony" <spam.DeleteThis@spam.com> wrote in message
>>> >news:npbeg39tfgj8aon14g1sc9ck6f3t9318nl@4ax.com...
>>> >> On Fri, 5 Oct 2007 15:17:57 +0000 (UTC), Andrew Smallshaw
>>> >> <andrews.DeleteThis@sdf.lonestar.org> wrote:
>>> >>
>>> >>> Yes, it did save money but why did the computer industry en masse decide
>>> >>> that it could safely be dispensed with?
>>> >>
>>> >> Because they don't particularly care if a customer's
>>> >> calculations/etc end up wrong if it's not guaranteed for
>>> >> some critical use. The industry didn't actually abandon it
>>> >> for critical uses.
>>> >
>>> >In fact, for critical usage some companies have gone far beyond normal
>>> >SEC/DED error correction in servers.
>>>
>>> If you are referring to ECC schemes and not memory mirroring, I'd be
>>> interested in examples that *don't* simply use n Hamming codewords to turn
>>> n-bit-wide full chip failures into correctable events - which really wasn't
>>> that remarkable, revolutionary or heroic when it was first employed - about 20
>>> years ago...
>>
>>I think you'll find it's been longer than 20 years. More likely
>>double that.
>
>I'll let you defend that statement with a cite Smile I'm sticking with the
>timeframe being the very early '80's, when Digital started shipping systems
>using x4 drams in volume and needed to survive a full chip failure...
>
ECC memory was used back in the core days, when "core" meant just that.
I remember ECC memory being available for S-100 bus machines; and it was
*old* technology even then.

You've always had to pay extra for it though.
The usual price/performance ratio is about 50% extra; and that hasn't
changed much over the many years; even though it should have with
today's prices on commodity things like memory set more by distribution
than complexity. That's why it doesn't cost nearly twice as much for
1-gig chips as it does for 500-meg chips these days.

>>> >I suppose I ought to trim the crossposting...
>>>
>>> And what fun would that be?
>>
>>Beat me to it.
>
>Bad habit of mine Wink
>
>/daytripper (meanwhile, I'm gonna go dangle flies in front of steelhead Smile

--
_____
/ ' / ™
,-/-, __ __. ____ /_
(_/ / (_(_/|_/ / <_/ <_
 >> Stay informed about: Questions about DDR RAM 
Back to top
Login to vote
Display posts from previous:   
Related Topics:
DDR Questions - I have a dell that has Dual Channel DDR PC3200 in it. Total of 256MB (2x128MB). I want to put another 512MB in it. I understand that it has to be installed in pairs to keep the memory working in dual channel. Does installing in pairs mean that I not....

Questions about SuperPower SP-586TX. - I have what appears to be a SuperPower SP-586TX (also known as San-Li & Hope Vision). BIOS ID = i430TX-W977TF-2A59ISMBC-00 Does anyone have a proper manual for this system? Also the clips that hold the DIMM in place are lower then what is on regul...

Parallel Port pin questions - I am trying to make my parallel port on my home computer drive a seven segment display. Most of it seems pretty simple, but I was wondering what to do with the pins I do not really need to perform this task..

XGI Volari duel GPU card + questions - Anyone who regularly keeps up with PC 3D graphics has already heard of the new Volari GPUs from XGI. Their highend Volari Duo solution will be a card with two DX9 GPUs in parallel. This has never been done before for *consumer* hardware using modern GPUs...

Misc Questions about Power supply - Hello, My question is about standard to fix and insert powersupply. What is Zippy ? I suppose a manufacturer but also a certain standard. What are standards to fix the PSU (I am not talking about connector) ? What is PFC, active or passive what ?..
   Hardware Problem Solving Community! (Home) -> Chips All times are: Pacific Time (US & Canada) (change)
Goto page Previous  1, 2, 3, 4, 5
Page 2 of 5

 
You can post new topics in this forum
You can reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum



[ Contact us | Terms of Service/Privacy Policy ]