Sign in with
Sign up | Sign in
Your question

Latest Athlon 64 product introductions

Last response: in CPUs
Share
Anonymous
a b à CPUs
June 3, 2004 4:52:44 AM

Archived from groups: comp.sys.ibm.pc.hardware.chips,comp.sys.intel (More info?)

AMD introduced 4 new Athlon 64 and FX models. However each one seems to be
designed for three different types of sockets. Athlon 64 3500+ and 3800+ are
designed for the Socket 939, while Athlon 64 3700+ is designed for Socket
754. Meanwhile are now two FX-53 models, the older one designed for Socket
940, while the newer one is designed for Socket 939. There's a table in this
article to help you keep it all clear in your mind:

http://www.eweek.com/article2/0,1759,1605309,00.asp

Yousuf Khan

--
Humans: contact me at ykhan at rogers dot com
Spambots: just reply to this email address ;-)
Anonymous
a b à CPUs
June 3, 2004 6:44:45 AM

Archived from groups: comp.sys.ibm.pc.hardware.chips,comp.sys.intel (More info?)

"Yousuf Khan" <bbbl67@ezrs.com> wrote in message
news:wDuvc.332025$0qd.165358@twister01.bloor.is.net.cable.rogers.com...
> AMD introduced 4 new Athlon 64 and FX models. However each one seems
to be
> designed for three different types of sockets. Athlon 64 3500+ and
3800+ are
> designed for the Socket 939, while Athlon 64 3700+ is designed for
Socket
> 754. Meanwhile are now two FX-53 models, the older one designed for
Socket
> 940, while the newer one is designed for Socket 939. There's a table
in this
> article to help you keep it all clear in your mind:
>
> http://www.eweek.com/article2/0,1759,1605309,00.asp

This is silly, how many socket types does one line of processors
require? Will there be a socket 941 next week?

Quote:
The FX-53 and 3500+, 3700+ and 3800+ include the Sunnyvale, Calif.,
company's Enhanced Virus Protection technology, which will be enabled
with Microsoft Corp.'s Windows XP Service Pack 2 release later this year
and is designed to add another layer of protection to desktops and
notebooks.
/Qoute

This is also absurd. CPU's should not be engineered to solve MS design
deficiencies.
Anonymous
a b à CPUs
June 3, 2004 6:44:46 AM

Archived from groups: comp.sys.ibm.pc.hardware.chips,comp.sys.intel (More info?)

Good point.

The more different "consumer" opterons (939/754) are released the
better the "real" opteron looks.

Really AMD is only following the market leader Intel. Intel and AMD
have had enough of folks running desktop processors for workstations
and/or servers. So therefore the best thing to do is engineer an
artifical difference. No more xp's running as mp's. AMD has finally
caught up with the p4/xeon duo. AMD has the amd64/opteron duo.

Whatever sounds confusing with all the models but still a lot better
than trying to decide which drug discount card to buy. Now that is a
real mess. ;-). Also a real ripoff.

Later



>>>>> "Anthony" == Anthony Fremont <spam@anywhere.com> writes:

Anthony> "Yousuf Khan" <bbbl67@ezrs.com> wrote in message
Anthony> news:wDuvc.332025$0qd.165358@twister01.bloor.is.net.cable.rogers.com...
>> AMD introduced 4 new Athlon 64 and FX models. However each one
>> seems
Anthony> to be
>> designed for three different types of sockets. Athlon 64 3500+ and
Anthony> 3800+ are
>> designed for the Socket 939, while Athlon 64 3700+ is designed for
Anthony> Socket
>> 754. Meanwhile are now two FX-53 models, the older one designed
>> for
Anthony> Socket
>> 940, while the newer one is designed for Socket 939. There's a
>> table
Anthony> in this
>> article to help you keep it all clear in your mind:
>>
>> http://www.eweek.com/article2/0,1759,1605309,00.asp

Anthony> This is silly, how many socket types does one line of
Anthony> processors require? Will there be a socket 941 next week?

Anthony> Quote: The FX-53 and 3500+, 3700+ and 3800+ include the
Anthony> Sunnyvale, Calif., company's Enhanced Virus Protection
Anthony> technology, which will be enabled with Microsoft Corp.'s
Anthony> Windows XP Service Pack 2 release later this year and is
Anthony> designed to add another layer of protection to desktops and
Anthony> notebooks. /Qoute

Anthony> This is also absurd. CPU's should not be engineered to solve
Anthony> MS design deficiencies.
Related resources
Anonymous
a b à CPUs
June 3, 2004 6:44:46 AM

Archived from groups: comp.sys.ibm.pc.hardware.chips,comp.sys.intel (More info?)

On Thu, 03 Jun 2004 02:44:45 GMT, "Anthony Fremont"
<spam@anywhere.com> wrote:
>"Yousuf Khan" <bbbl67@ezrs.com> wrote in message
>news:wDuvc.332025$0qd.165358@twister01.bloor.is.net.cable.rogers.com...
>> AMD introduced 4 new Athlon 64 and FX models. However each one seems
>to be
>> designed for three different types of sockets. Athlon 64 3500+ and
>3800+ are
>> designed for the Socket 939, while Athlon 64 3700+ is designed for
>Socket
>> 754. Meanwhile are now two FX-53 models, the older one designed for
>Socket
>> 940, while the newer one is designed for Socket 939. There's a table
>in this
>> article to help you keep it all clear in your mind:
>>
>> http://www.eweek.com/article2/0,1759,1605309,00.asp
>
>This is silly, how many socket types does one line of processors
>require? Will there be a socket 941 next week?

This has been discussed before, but to sum up briefly:

- Originally all Athlon64 processors were to be socket 754, that's it,
that's all. Socket 940 would be used for all Opteron processors and
never the two shall meet

- Then a few months before the Athlon64 was released, dual-channel
memory became a checkmark feature that all high-end systems needed to
have. Socket 754 wouldn't cut it and AMD needed something quick.
Easy solution: take an Opteron and sell it as a desktop chip in socket
940.

- For the long-term though, AMD needed a proper solution, one designed
for dual-channel memory for the desktop. The result is socket 939.

That should be it for sockets for the entire K8 line-up. 3 sockets
actually isn't all that bad, especially if you compare it to Intel's
P4/Xeon line. They started with Socket 423 for the desktop and Socket
603 for workstations/server. Then moved to Socket 474 for the
desktop/laptop and socket 604 for workstations/servers. In another
couple months they'll move to a totally new socket 775 for desktops
(and laptops?), and they may have another new socket for servers soon.

The PIII was even worse though, since it actually had three different
and not-quite-compatible Socket 370s as well as Slot 1, not to mention
Slot 2 for servers. AMD's old Athlon chips also went through a few
revisions, first going from Slot A to Socket A and then changing the
voltage and bus speed sufficiently that old socket A boards couldn't
handle new chips.

Long story short, socket changes are the norm, not the exception. AMD
having only 3 sockets for a processor core isn't too bad.

>Quote:
>The FX-53 and 3500+, 3700+ and 3800+ include the Sunnyvale, Calif.,
>company's Enhanced Virus Protection technology, which will be enabled
>with Microsoft Corp.'s Windows XP Service Pack 2 release later this year
>and is designed to add another layer of protection to desktops and
>notebooks.
>/Qoute
>
>This is also absurd. CPU's should not be engineered to solve MS design
>deficiencies.

Again, this has been discussed at length, but the short version of
this is that it's a feature that *SHOULD* have been in the processor
long ago. x86 was one of a very small number of high-end
architectures that doesn't already have this feature (except in
segmentation, and there's already a big flame-war going on about that
is this newsgroup!).

It's not just for Windows either, Linux and some of the BSD's already
make use of this security feature in their x86-64 distributions. Of
course, calling it "Virus Protection" is a total misnomer, it won't do
a thing to stop viruses. What it will help prevent is worms and
hacking attempts through buffer overruns. However the difference
between a "virus" and a "worm" has been pretty much lost on the
mainstream media, let along the general public.

-------------
Tony Hill
hilla <underscore> 20 <at> yahoo <dot> ca
Anonymous
a b à CPUs
June 3, 2004 11:50:05 AM

Archived from groups: comp.sys.ibm.pc.hardware.chips,comp.sys.intel (More info?)

Tony Hill <hilla_nospam_20@yahoo.ca> wrote:

>Long story short, socket changes are the norm, not the exception. AMD
>having only 3 sockets for a processor core isn't too bad.

(so far)

8)
June 3, 2004 1:07:54 PM

Archived from groups: comp.sys.ibm.pc.hardware.chips,comp.sys.intel (More info?)

"Anthony Fremont" <spam@anywhere.com> wrote :

> Quote:
> The FX-53 and 3500+, 3700+ and 3800+ include the Sunnyvale, Calif.,
> company's Enhanced Virus Protection technology, which will be
> enabled with Microsoft Corp.'s Windows XP Service Pack 2 release
> later this year and is designed to add another layer of protection
> to desktops and notebooks.
> /Qoute
>
> This is also absurd. CPU's should not be engineered to solve MS
> design deficiencies.

this is NX bit, some stupid marketoid turned it into "av protection tech
bla bla, and now we rule baby" "/


Pozdrawiam.
--
RusH //
http://pulse.pdi.net/~rush/qv30/
Like ninjas, true hackers are shrouded in secrecy and mystery.
You may never know -- UNTIL IT'S TOO LATE.
Anonymous
a b à CPUs
June 3, 2004 2:48:51 PM

Archived from groups: comp.sys.ibm.pc.hardware.chips,comp.sys.intel (More info?)

"Tony Hill" <hilla_nospam_20@yahoo.ca> wrote in message

> Again, this has been discussed at length, but the short version of
> this is that it's a feature that *SHOULD* have been in the processor
> long ago. x86 was one of a very small number of high-end
> architectures that doesn't already have this feature (except in
> segmentation, and there's already a big flame-war going on about that
> is this newsgroup!).
>
> It's not just for Windows either, Linux and some of the BSD's already
> make use of this security feature in their x86-64 distributions. Of
> course, calling it "Virus Protection" is a total misnomer, it won't do
> a thing to stop viruses. What it will help prevent is worms and
> hacking attempts through buffer overruns. However the difference
> between a "virus" and a "worm" has been pretty much lost on the
> mainstream media, let along the general public.

In light of that, I agree with you. Rather than taking the comment at
face value, I should have looked to see what was really being added to
the proc. I know better than to listen to media hype, but I guess I got
caught with my pants down on that one by taking their comment at face
value. Segment protection features in the CPU as you described are
sensible.
Anonymous
a b à CPUs
June 3, 2004 6:32:38 PM

Archived from groups: comp.sys.ibm.pc.hardware.chips,comp.sys.intel (More info?)

"RusH" <logistyka1@pf.pl> wrote in message
news:Xns94FD705D0E272RusHcomputersystems@193.110.122.80...
> "Anthony Fremont" <spam@anywhere.com> wrote :

> >This is also absurd. CPU's should not be engineered to solve MS
> > design deficiencies.
>
> this is NX bit, some stupid marketoid turned it into "av protection
tech
> bla bla, and now we rule baby" "/

Ahh....ok. That's allot more sensible than "antivirus protection". NX
is a good idea, data pages shouldn't be executable and code pages
shouldn't be writable (well most of the time anyway ;-).
Anonymous
a b à CPUs
June 3, 2004 10:07:42 PM

Archived from groups: comp.sys.ibm.pc.hardware.chips,comp.sys.intel (More info?)

Anthony Fremont <spam@anywhere.com> wrote:
> This is silly, how many socket types does one line of processors
> require? Will there be a socket 941 next week?

There were originally only two sockets available, Socket 754 and 940. Socket
754 was for single-channel, unbuffered DDR RAM; it was only going to be used
by Athlon 64 systems. Socket 940 was dual-channel, buffered DDR RAM; it was
only to be used by Opteron systems. But then it introduced the Athlon 64 FX
which was really a rebadged Opteron. Thus Athlon 64 FX and Opteron were
sharing the same type of motherboard platform, so the distinction between
Opteron and Athlon 64 blurred.

Later AMD decided that it would like Athlon 64 systems to be dual-channel
too like the Opterons, but it didn't want them to use the expensive buffered
RAM like the Opterons. So Socket 939 was born -- dual-channel, unbuffered
DDR RAM.

> Quote:
> The FX-53 and 3500+, 3700+ and 3800+ include the Sunnyvale, Calif.,
> company's Enhanced Virus Protection technology, which will be enabled
> with Microsoft Corp.'s Windows XP Service Pack 2 release later this
> year and is designed to add another layer of protection to desktops
> and notebooks.
> /Qoute
>
> This is also absurd. CPU's should not be engineered to solve MS
> design deficiencies.

Well, "anti-virus" is really a marketing term. It's really a type of
anti-buffer overflow feature, and it's not really something meant to be used
against viruses, but against worms.

Yousuf Khan
Anonymous
a b à CPUs
June 4, 2004 12:24:54 AM

Archived from groups: comp.sys.ibm.pc.hardware.chips,comp.sys.intel (More info?)

>>>>> "Yousuf" == Yousuf Khan <bbbl67@ezrs.com> writes:

Yousuf> and 940. Socket 754 was for single-channel, unbuffered DDR
Yousuf> RAM; it was only going to be used by Athlon 64 systems.
Yousuf> Socket 940 was dual-channel, buffered DDR RAM; it was only to
Yousuf> be used by Opteron systems. But then it introduced the Athlon
Yousuf> 64 FX which was really a rebadged Opteron. Thus Athlon 64 FX
Yousuf> and Opteron were sharing the same type of motherboard
Yousuf> platform, so the distinction between Opteron and Athlon 64
Yousuf> blurred.

A quick question if one believes the recent benchmarks posted at
several sites it appears the performance advantage of dual-channel
memory over single-channel is really not significant. With faster
processors or faster memory will dual-channel memory make significant
performance gains over single-channel memory in the future? I guess
does the market expect dual channel because that appears to be the
standard?

Thanks
Anonymous
a b à CPUs
June 4, 2004 12:54:04 AM

Archived from groups: comp.sys.ibm.pc.hardware.chips,comp.sys.intel (More info?)

Yousuf Khan wrote:

> Anthony Fremont <spam@anywhere.com> wrote:
>
>>This is silly, how many socket types does one line of processors
>>require? Will there be a socket 941 next week?
>
>
> There were originally only two sockets available, Socket 754 and 940. Socket
> 754 was for single-channel, unbuffered DDR RAM; it was only going to be used
> by Athlon 64 systems. Socket 940 was dual-channel, buffered DDR RAM; it was
> only to be used by Opteron systems. But then it introduced the Athlon 64 FX
> which was really a rebadged Opteron. Thus Athlon 64 FX and Opteron were
> sharing the same type of motherboard platform, so the distinction between
> Opteron and Athlon 64 blurred.
>
> Later AMD decided that it would like Athlon 64 systems to be dual-channel
> too like the Opterons, but it didn't want them to use the expensive buffered
> RAM like the Opterons. So Socket 939 was born -- dual-channel, unbuffered
> DDR RAM.
>

There is no "later" about it. AMD planned for sockets 754, 939
and 940 right from day one. Manufacturers were eager to start
making socket 939 boards a full year ago - there just weren't
any Socket 939 chips. No point in making the boards if the
consumers can't buy chips to put in them. And now that the 939
chips are finally starting to come out, manufacturers seem to
have lost a lot of their initial enthusiasm for Socket 939 -
probably due in part to the success of Socket 754. They seem
to be happy with 754 for now and seem to be planning on tying
939 together with PCI-E for the next generation of desktops
rather than for today.
Anonymous
a b à CPUs
June 4, 2004 8:09:29 AM

Archived from groups: comp.sys.ibm.pc.hardware.chips,comp.sys.intel (More info?)

On Thu, 03 Jun 2004 20:24:54 -0500, Post Replies Here Please
<spamme@edge.net> wrote:
>>>>>> "Yousuf" == Yousuf Khan <bbbl67@ezrs.com> writes:
>
> Yousuf> and 940. Socket 754 was for single-channel, unbuffered DDR
> Yousuf> RAM; it was only going to be used by Athlon 64 systems.
> Yousuf> Socket 940 was dual-channel, buffered DDR RAM; it was only to
> Yousuf> be used by Opteron systems. But then it introduced the Athlon
> Yousuf> 64 FX which was really a rebadged Opteron. Thus Athlon 64 FX
> Yousuf> and Opteron were sharing the same type of motherboard
> Yousuf> platform, so the distinction between Opteron and Athlon 64
> Yousuf> blurred.
>
>A quick question if one believes the recent benchmarks posted at
>several sites it appears the performance advantage of dual-channel
>memory over single-channel is really not significant.

This is not altogether surprising, though the difference does exist at
least. This page on Ace's Hardware has a good run-down of how much
dual-channel memory helps:

http://www.aceshardware.com/read.jsp?id=65000314

Quick summary: performance increases anywhere from 0% up to 23.8%, but
with the average being about 3.9% for the 28 benchmarks they ran.
Nothing huge, but it is worth about a speed grade of a processor.

Another nice advantage of dual-channel Athlon64 systems is that they
allow more memory to be installed. The single-channel Socket 754
Athlon64 chips could only handle a maximum of 3 DIMMs, and many boards
either didn't have the third DIMM slot or they weren't reliable if all
3 DIMM slots were used. The new dual-channel Socket 939 Athlon64
systems can handle up to 4 DIMMs reliably (though usually with
somewhat slower memory timings).

> With faster
>processors or faster memory will dual-channel memory make significant
>performance gains over single-channel memory in the future?

Not likely. Latency is most important for most applications,
bandwidth is somewhat secondary.

In fact, despite populate belief, dual-channel memory doesn't help
performance for Intel chips much either and they are already running
at 3GHz+. Take a look sometime at a comparison between Intel's 848
(single channel) and otherwise identical 865PE (dual-channel)
chipsets. Performance difference is a bit larger than for these AMD
chips, but typically only in the 4-5% range.

A lot of people thought that P4 chips were super-bandwidth hungry
because of early comparison between the single-channel 845 chipset and
the dual-channel 875 chipset, but now it seems that most of the
performance difference came from factors other than just memory
bandwidth (perhaps most importantly is that Intel did a bang-up job in
getting super-low memory latency on the i875 chipset).

> I guess
>does the market expect dual channel because that appears to be the
>standard?

That's pretty much it. Intel has dual-channel, therefore AMD should
have dual-channel as well, right?

-------------
Tony Hill
hilla <underscore> 20 <at> yahoo <dot> ca
Anonymous
a b à CPUs
June 4, 2004 8:51:05 AM

Archived from groups: comp.sys.ibm.pc.hardware.chips,comp.sys.intel (More info?)

Post Replies Here Please <spamme@edge.net> writes:

>>>>>> "Yousuf" == Yousuf Khan <bbbl67@ezrs.com> writes:
>
> Yousuf> and 940. Socket 754 was for single-channel, unbuffered DDR
> Yousuf> RAM; it was only going to be used by Athlon 64 systems.
> Yousuf> Socket 940 was dual-channel, buffered DDR RAM; it was only to
> Yousuf> be used by Opteron systems. But then it introduced the Athlon
> Yousuf> 64 FX which was really a rebadged Opteron. Thus Athlon 64 FX
> Yousuf> and Opteron were sharing the same type of motherboard
> Yousuf> platform, so the distinction between Opteron and Athlon 64
> Yousuf> blurred.
>
> A quick question if one believes the recent benchmarks posted at
> several sites it appears the performance advantage of dual-channel
> memory over single-channel is really not significant.

This is quite believable. Theoretically, dual-channel memory delivers
twice the _peak_ memory bandwidth. In real life, however, the
performance gain is lowered due to the latency of the memories are
incurred almost every time the processor makes a jump in the
addressing to the RAM.

So streaming memory bandwidth tests will show you x2 the bandwidth,
but it is very seldom this is obtainable for your joe-average program.

> With faster
> processors or faster memory will dual-channel memory make significant
> performance gains over single-channel memory in the future? I guess
> does the market expect dual channel because that appears to be the
> standard?

I haven't done any research on this, but I'd expect that the built-in
memory controller to have a bigger impact than single/dual-channel
memory.

Regards,


Kai
Anonymous
a b à CPUs
June 4, 2004 10:45:41 PM

Archived from groups: comp.sys.ibm.pc.hardware.chips,comp.sys.intel (More info?)

Post Replies Here Please <spamme@edge.net> wrote:
> A quick question if one believes the recent benchmarks posted at
> several sites it appears the performance advantage of dual-channel
> memory over single-channel is really not significant. With faster
> processors or faster memory will dual-channel memory make significant
> performance gains over single-channel memory in the future? I guess
> does the market expect dual channel because that appears to be the
> standard?

I think AMD might be able save some money by going dual-channel. It looks
like the benchmarks show that you get pretty identical performance if you
halve the cache, but go dual-channel. Reducing the cache would help the
price of these things.

Yousuf Khan
Anonymous
a b à CPUs
June 5, 2004 12:19:03 AM

Archived from groups: comp.sys.ibm.pc.hardware.chips,comp.sys.intel (More info?)

On Fri, 04 Jun 2004 04:09:29 -0400, Tony Hill <hilla_nospam_20@yahoo.ca>
wrote:

>Another nice advantage of dual-channel Athlon64 systems is that they
>allow more memory to be installed. The single-channel Socket 754
>Athlon64 chips could only handle a maximum of 3 DIMMs, and many boards
>either didn't have the third DIMM slot or they weren't reliable if all
>3 DIMM slots were used.

Dunno who's rule it is but the recent MSI K8T Neo mbrd I did with Athlon64
does not support the 3 DIMMs at DDR400 - right there in the manual. That's
with unbuffered DIMMs of course.

Rgds, George Macdonald

"Just because they're paranoid doesn't mean you're not psychotic" - Who, me??
Anonymous
a b à CPUs
June 6, 2004 4:44:27 PM

Archived from groups: comp.sys.ibm.pc.hardware.chips,comp.sys.intel (More info?)

Thanks to all for the interesting discussion about dual channel vs.
single channel memory. Also thanks for the link.

Alan
June 8, 2004 1:19:08 PM

Archived from groups: comp.sys.ibm.pc.hardware.chips,comp.sys.intel (More info?)

On Thu, 03 Jun 2004 18:07:42 GMT, "Yousuf Khan" <bbbl67@ezrs.com> wrote:

>
>Later AMD decided that it would like Athlon 64 systems to be dual-channel
>too like the Opterons, but it didn't want them to use the expensive buffered
>RAM like the Opterons. So Socket 939 was born -- dual-channel, unbuffered
>DDR RAM.

939 boards are also cheaper to make, 4 layer instead of 6.
Anonymous
a b à CPUs
June 8, 2004 8:37:02 PM

Archived from groups: comp.sys.ibm.pc.hardware.chips,comp.sys.intel (More info?)

Ed wrote:

> On Thu, 03 Jun 2004 18:07:42 GMT, Yousuf Khan wrote:
>
>> Later AMD decided that it would like Athlon 64 systems to be
>> dual-channel too like the Opterons, but it didn't want them to use
>> the expensive buffered RAM like the Opterons. So Socket 939 was
>> born -- dual-channel, unbuffered DDR RAM.
>
>
> 939 boards are also cheaper to make, 4 layer instead of 6.

Can somebody explain why socket 754 motherboards needed two more layers?
June 8, 2004 8:37:03 PM

Archived from groups: comp.sys.ibm.pc.hardware.chips,comp.sys.intel (More info?)

On Tue, 08 Jun 2004 16:37:02 +0200, Grumble <a@b.c> wrote:

>Ed wrote:
>
>> On Thu, 03 Jun 2004 18:07:42 GMT, Yousuf Khan wrote:
>>
>>> Later AMD decided that it would like Athlon 64 systems to be
>>> dual-channel too like the Opterons, but it didn't want them to use
>>> the expensive buffered RAM like the Opterons. So Socket 939 was
>>> born -- dual-channel, unbuffered DDR RAM.
>>
>>
>> 939 boards are also cheaper to make, 4 layer instead of 6.
>
>Can somebody explain why socket 754 motherboards needed two more layers?


940 is 6
754 are 4 layer isn't it?
Anonymous
a b à CPUs
June 8, 2004 10:00:29 PM

Archived from groups: comp.sys.ibm.pc.hardware.chips,comp.sys.intel (More info?)

Ed <uranidiot@behappy.com> wrote:
> On Tue, 08 Jun 2004 16:37:02 +0200, Grumble <a@b.c> wrote:
>
>> Ed wrote:
>>> 939 boards are also cheaper to make, 4 layer instead of 6.
>>
>> Can somebody explain why socket 754 motherboards needed two more
>> layers?
>
>
> 940 is 6
> 754 are 4 layer isn't it?

Oh geez, now we got two Ed's, and they're gonna carry on a confusing debate
amongst themselves. :-)

Yousuf Khan
Anonymous
a b à CPUs
June 8, 2004 10:22:41 PM

Archived from groups: comp.sys.ibm.pc.hardware.chips,comp.sys.intel (More info?)

You know what they say,

"two Eds are better than one"


"Yousuf Khan" <bbbl67@ezrs.com> wrote in message
news:19nxc.118693$Ar.81900@twister01.bloor.is.net.cable.rogers.com...
> Ed <uranidiot@behappy.com> wrote:
> > On Tue, 08 Jun 2004 16:37:02 +0200, Grumble <a@b.c> wrote:
> >
> >> Ed wrote:
> >>> 939 boards are also cheaper to make, 4 layer instead of 6.
> >>
> >> Can somebody explain why socket 754 motherboards needed two more
> >> layers?
> >
> >
> > 940 is 6
> > 754 are 4 layer isn't it?
>
> Oh geez, now we got two Ed's, and they're gonna carry on a confusing
debate
> amongst themselves. :-)
>
> Yousuf Khan
>
>
Anonymous
a b à CPUs
June 9, 2004 2:04:52 AM

Archived from groups: comp.sys.ibm.pc.hardware.chips,comp.sys.intel (More info?)

In article <8libc09361ibdvdja5icpc9s952hq4eefi@4ax.com>,
nospam@email.com says...
> On Thu, 03 Jun 2004 18:07:42 GMT, "Yousuf Khan" <bbbl67@ezrs.com> wrote:
>
> >
> >Later AMD decided that it would like Athlon 64 systems to be dual-channel
> >too like the Opterons, but it didn't want them to use the expensive buffered
> >RAM like the Opterons. So Socket 939 was born -- dual-channel, unbuffered
> >DDR RAM.
>
> 939 boards are also cheaper to make, 4 layer instead of 6.

Why? I don't see the memory differences as significant.

--
Keith
Anonymous
a b à CPUs
June 9, 2004 2:06:02 AM

Archived from groups: comp.sys.ibm.pc.hardware.chips,comp.sys.intel (More info?)

In article <19nxc.118693$Ar.81900
@twister01.bloor.is.net.cable.rogers.com>, bbbl67@ezrs.com
says...
> Ed <uranidiot@behappy.com> wrote:
> > On Tue, 08 Jun 2004 16:37:02 +0200, Grumble <a@b.c> wrote:
> >
> >> Ed wrote:
> >>> 939 boards are also cheaper to make, 4 layer instead of 6.
> >>
> >> Can somebody explain why socket 754 motherboards needed two more
> >> layers?
> >
> >
> > 940 is 6
> > 754 are 4 layer isn't it?
>
> Oh geez, now we got two Ed's, and they're gonna carry on a confusing debate
> amongst themselves. :-)

Nope. Same Ed, just a different moniker (look at the headings).

--
Keith
Anonymous
a b à CPUs
June 9, 2004 9:22:38 AM

Archived from groups: comp.sys.ibm.pc.hardware.chips,comp.sys.intel (More info?)

KR Williams <krw@att.biz> wrote:
> In article <19nxc.118693$Ar.81900
> @twister01.bloor.is.net.cable.rogers.com>, bbbl67@ezrs.com
> says...
>> Ed <uranidiot@behappy.com> wrote:
>>> On Tue, 08 Jun 2004 16:37:02 +0200, Grumble <a@b.c> wrote:
>>>
>>>> Ed wrote:
>>>>> 939 boards are also cheaper to make, 4 layer instead of 6.
>>>>
>>>> Can somebody explain why socket 754 motherboards needed two more
>>>> layers?
>>>
>>>
>>> 940 is 6
>>> 754 are 4 layer isn't it?
>>
>> Oh geez, now we got two Ed's, and they're gonna carry on a confusing
>> debate amongst themselves. :-)
>
> Nope. Same Ed, just a different moniker (look at the headings).

If they are the same Ed, then why is one asking a question of the other?
This has the potential of making the Tony vs. Tony Hill debates look
comprehensible.

Yousuf Khan
Anonymous
a b à CPUs
June 9, 2004 7:44:31 PM

Archived from groups: comp.sys.ibm.pc.hardware.chips,comp.sys.intel (More info?)

KR Williams wrote:
> In article <8libc09361ibdvdja5icpc9s952hq4eefi@4ax.com>,
> nospam@email.com says...
>
>>On Thu, 03 Jun 2004 18:07:42 GMT, "Yousuf Khan" <bbbl67@ezrs.com> wrote:
>>
>>
>>>Later AMD decided that it would like Athlon 64 systems to be dual-channel
>>>too like the Opterons, but it didn't want them to use the expensive buffered
>>>RAM like the Opterons. So Socket 939 was born -- dual-channel, unbuffered
>>>DDR RAM.
>>
>>939 boards are also cheaper to make, 4 layer instead of 6.
>
>
> Why? I don't see the memory differences as significant.

Same question crossed my mind. My guess is that they weren't *too*
bothered about Opteron boards being costly so they didn't put much
effort into keeping the layer count down. I'll bet that some bright
spark spotted an alternative pin-out that would allow 4 layer
boards sometime after S940 was released, hence the magic new S939
that allows 4-Layer boards. :) 

Cheers,
Rupert
Anonymous
a b à CPUs
June 9, 2004 7:44:32 PM

Archived from groups: comp.sys.ibm.pc.hardware.chips,comp.sys.intel (More info?)

Rupert Pigott wrote:

> KR Williams wrote:
>
>> In article <8libc09361ibdvdja5icpc9s952hq4eefi@4ax.com>,
>> nospam@email.com says...
>>
>>> On Thu, 03 Jun 2004 18:07:42 GMT, "Yousuf Khan" <bbbl67@ezrs.com> wrote:
>>>
>>>
>>>> Later AMD decided that it would like Athlon 64 systems to be
>>>> dual-channel
>>>> too like the Opterons, but it didn't want them to use the expensive
>>>> buffered
>>>> RAM like the Opterons. So Socket 939 was born -- dual-channel,
>>>> unbuffered
>>>> DDR RAM.
>>>
>>>
>>> 939 boards are also cheaper to make, 4 layer instead of 6.
>>
>>
>>
>> Why? I don't see the memory differences as significant.
>
>
> Same question crossed my mind. My guess is that they weren't *too*
> bothered about Opteron boards being costly so they didn't put much
> effort into keeping the layer count down. I'll bet that some bright
> spark spotted an alternative pin-out that would allow 4 layer
> boards sometime after S940 was released, hence the magic new S939
> that allows 4-Layer boards. :) 
>

No. Socket 940, 939, and 754 were all announced at the same time.
Manufacturers were ready to start cranking out Socket 939 boards
more than a year ago but AMD just wasn't producing the chips.

As well, PDF illustrating the pin-outs for all three sockets were
available at AMD's site early last year and I would expect they
are still there if you are interested.
Anonymous
a b à CPUs
June 9, 2004 8:24:10 PM

Archived from groups: comp.sys.ibm.pc.hardware.chips,comp.sys.intel (More info?)

Yousuf Khan wrote:

> KR Williams <krw@att.biz> wrote:
>> In article <19nxc.118693$Ar.81900
>> @twister01.bloor.is.net.cable.rogers.com>, bbbl67@ezrs.com
>> says...
>>> Ed <uranidiot@behappy.com> wrote:
>>>> On Tue, 08 Jun 2004 16:37:02 +0200, Grumble <a@b.c> wrote:
>>>>
>>>>> Ed wrote:
>>>>>> 939 boards are also cheaper to make, 4 layer instead of 6.
>>>>>
>>>>> Can somebody explain why socket 754 motherboards needed two
>>>>> more layers?
>>>>
>>>>
>>>> 940 is 6
>>>> 754 are 4 layer isn't it?
>>>
>>> Oh geez, now we got two Ed's, and they're gonna carry on a
>>> confusing debate amongst themselves. :-)
>>
>> Nope. Same Ed, just a different moniker (look at the headings).
>
> If they are the same Ed, then why is one asking a question of the
> other?

My news server has a "Grumble" post threaded in between.

> This has the potential of making the Tony vs. Tony Hill
> debates look comprehensible.

;-) That one got me!

--
Keith
Anonymous
a b à CPUs
June 9, 2004 8:26:11 PM

Archived from groups: comp.sys.ibm.pc.hardware.chips,comp.sys.intel (More info?)

Rupert Pigott wrote:

> KR Williams wrote:
>> In article <8libc09361ibdvdja5icpc9s952hq4eefi@4ax.com>,
>> nospam@email.com says...
>>
>>>On Thu, 03 Jun 2004 18:07:42 GMT, "Yousuf Khan" <bbbl67@ezrs.com>
>>>wrote:
>>>
>>>
>>>>Later AMD decided that it would like Athlon 64 systems to be
>>>>dual-channel too like the Opterons, but it didn't want them to
>>>>use the expensive buffered RAM like the Opterons. So Socket 939
>>>>was born -- dual-channel, unbuffered DDR RAM.
>>>
>>>939 boards are also cheaper to make, 4 layer instead of 6.
>>
>>
>> Why? I don't see the memory differences as significant.
>
> Same question crossed my mind. My guess is that they weren't *too*
> bothered about Opteron boards being costly so they didn't put much
> effort into keeping the layer count down. I'll bet that some
> bright spark spotted an alternative pin-out that would allow 4
> layer boards sometime after S940 was released, hence the magic new
> S939 that allows 4-Layer boards. :) 

The only possible difference I can see is that the S939 boards won't
be pretending to do more than one processor, so may be somewhat
simpler that way. However that would assume the HT links aren't
being used for anything else. So why all the pins?

--
Keith
Anonymous
a b à CPUs
June 9, 2004 8:46:20 PM

Archived from groups: comp.sys.ibm.pc.hardware.chips,comp.sys.intel (More info?)

Rob Stow wrote:
> Rupert Pigott wrote:
>
>> KR Williams wrote:
>>
>>> In article <8libc09361ibdvdja5icpc9s952hq4eefi@4ax.com>,
>>> nospam@email.com says...
>>>
>>>> On Thu, 03 Jun 2004 18:07:42 GMT, "Yousuf Khan" <bbbl67@ezrs.com>
>>>> wrote:
>>>>
>>>>
>>>>> Later AMD decided that it would like Athlon 64 systems to be
>>>>> dual-channel
>>>>> too like the Opterons, but it didn't want them to use the expensive
>>>>> buffered
>>>>> RAM like the Opterons. So Socket 939 was born -- dual-channel,
>>>>> unbuffered
>>>>> DDR RAM.
>>>>
>>>>
>>>>
>>>> 939 boards are also cheaper to make, 4 layer instead of 6.
>>>
>>>
>>>
>>>
>>> Why? I don't see the memory differences as significant.
>>
>>
>>
>> Same question crossed my mind. My guess is that they weren't *too*
>> bothered about Opteron boards being costly so they didn't put much
>> effort into keeping the layer count down. I'll bet that some bright
>> spark spotted an alternative pin-out that would allow 4 layer
>> boards sometime after S940 was released, hence the magic new S939
>> that allows 4-Layer boards. :) 
>>
>
> No. Socket 940, 939, and 754 were all announced at the same time.
> Manufacturers were ready to start cranking out Socket 939 boards
> more than a year ago but AMD just wasn't producing the chips.

Eeek, I hadn't twigged 939 & 940 were launched at the same time. Seems
a bit odd that the 4/6 layer difference exists in that case.

> As well, PDF illustrating the pin-outs for all three sockets were
> available at AMD's site early last year and I would expect they
> are still there if you are interested.

Must admit I'm curious to see what kind of differences exist between 939
and 940.

Thanks,
Rupert
Anonymous
a b à CPUs
June 9, 2004 8:48:23 PM

Archived from groups: comp.sys.ibm.pc.hardware.chips,comp.sys.intel (More info?)

Bitstring <1086792271.200887@teapot.planet.gong>, from the wonderful
person Rupert Pigott <roo@try-removing-this.darkboong.demon.co.uk> said
<snip>
>>>939 boards are also cheaper to make, 4 layer instead of 6.

>> Why? I don't see the memory differences as significant.
>
>Same question crossed my mind. My guess is that they weren't *too*
>bothered about Opteron boards being costly so they didn't put much
>effort into keeping the layer count down. I'll bet that some bright
>spark spotted an alternative pin-out that would allow 4 layer
>boards sometime after S940 was released, hence the magic new S939
>that allows 4-Layer boards. :) 

I'm not sure you can blame the pinout - I can't see why you can't route
anything with 4 layers (actually, iirc, with 2 .. in theory, for
constant width wires), unless/until you start needing massive power
distribution and ground planes?

--
GSV Three Minds in a Can
Outgoing Msgs are Turing Tested,and indistinguishable from human typing.
Anonymous
a b à CPUs
June 9, 2004 8:48:24 PM

Archived from groups: comp.sys.ibm.pc.hardware.chips,comp.sys.intel (More info?)

GSV Three Minds in a Can wrote:

> Bitstring <1086792271.200887@teapot.planet.gong>, from the
> wonderful person Rupert Pigott
> <roo@try-removing-this.darkboong.demon.co.uk> said <snip>
>>>>939 boards are also cheaper to make, 4 layer instead of 6.
>
>>> Why? I don't see the memory differences as significant.
>>
>>Same question crossed my mind. My guess is that they weren't *too*
>>bothered about Opteron boards being costly so they didn't put much
>>effort into keeping the layer count down. I'll bet that some
>>bright spark spotted an alternative pin-out that would allow 4
>>layer boards sometime after S940 was released, hence the magic new
>>S939 that allows 4-Layer boards. :) 
>
> I'm not sure you can blame the pinout - I can't see why you can't
> route anything with 4 layers (actually, iirc, with 2 .. in theory,
> for constant width wires), unless/until you start needing massive
> power distribution and ground planes?

Read that last sentence again. ;-)

The reality is that you need the ground/power planes for electrical
reasons, other than the *MASSIVE* power these things dissipate.
Think about 60W at ~1.25V. That's a wee bit of current. In
addition the impedance of the power distribution must be kept as
low as possible to reduce noise. The planes are also necessary to
keep the impedance of the signal lines constant. Two of the four
planes were power planes in the 486 days, without the massive
currents we now see. Thus, a four-layer board has only two wiring
planes, which are normally wired horizontally and vertically.
"Manhattan" wiring (streets on one side, avenues on the other) to
hook everything together.

Yes, you can theoretically wire anything using two layers, with an
infinite wiring space, infinitely long wires, and an infinite
number of vias. To make everything work, there are restrictions on
all of these as well as the differences in these from one wire to
another.

--
Keith
Anonymous
a b à CPUs
June 9, 2004 11:07:07 PM

Archived from groups: comp.sys.ibm.pc.hardware.chips,comp.sys.intel (More info?)

Rupert Pigott <roo@try-removing-this.darkboong.demon.co.uk> wrote:
> Must admit I'm curious to see what kind of differences exist between
> 939 and 940.

From a purely end-user perspective, the biggest difference between 939 and
940 is that 939 is for unbuffered DDR, while 940 is for server-class
buffered DDR. I assume that since 940 uses buffered DDR, the potential
exists to outfit 940 board with well over two DIMMs per processor.

Yousuf Khan
Anonymous
a b à CPUs
June 9, 2004 11:07:08 PM

Archived from groups: comp.sys.ibm.pc.hardware.chips,comp.sys.intel (More info?)

Yousuf Khan wrote:

> Rupert Pigott <roo@try-removing-this.darkboong.demon.co.uk> wrote:
>> Must admit I'm curious to see what kind of differences exist
>> between 939 and 940.
>
> From a purely end-user perspective, the biggest difference between
> 939 and 940 is that 939 is for unbuffered DDR, while 940 is for
> server-class buffered DDR. I assume that since 940 uses buffered
> DDR, the potential exists to outfit 940 board with well over two
> DIMMs per processor.

But, but... Why would a 940 board that *only* supports four DIMMs
be more complicated than a 939 board that only supports 4 DIMMs?
DIMMs is Dimms (at least on this level).

--
Keith
Anonymous
a b à CPUs
June 10, 2004 2:09:17 AM

Archived from groups: comp.sys.ibm.pc.hardware.chips (More info?)

Bitstring <3oadnSFRhc3F6FrdRVn-vA@adelphia.com>, from the wonderful
person K Williams <krw@adelphia.net> said
>GSV Three Minds in a Can wrote:
>
>> Bitstring <1086792271.200887@teapot.planet.gong>, from the
>> wonderful person Rupert Pigott
>> <roo@try-removing-this.darkboong.demon.co.uk> said <snip>
>>>>>939 boards are also cheaper to make, 4 layer instead of 6.
>>
>>>> Why? I don't see the memory differences as significant.
>>>
>>>Same question crossed my mind. My guess is that they weren't *too*
>>>bothered about Opteron boards being costly so they didn't put much
>>>effort into keeping the layer count down. I'll bet that some
>>>bright spark spotted an alternative pin-out that would allow 4
>>>layer boards sometime after S940 was released, hence the magic new
>>>S939 that allows 4-Layer boards. :) 
>>
>> I'm not sure you can blame the pinout - I can't see why you can't
>> route anything with 4 layers (actually, iirc, with 2 .. in theory,
>> for constant width wires), unless/until you start needing massive
>> power distribution and ground planes?
>
>Read that last sentence again. ;-)
>
>The reality is that you need the ground/power planes for electrical
>reasons, other than the *MASSIVE* power these things dissipate.
>Think about 60W at ~1.25V. That's a wee bit of current. In
>addition the impedance of the power distribution must be kept as
>low as possible to reduce noise. The planes are also necessary to
>keep the impedance of the signal lines constant. Two of the four
>planes were power planes in the 486 days, without the massive
>currents we now see.

>Yes, you can theoretically wire anything using two layers, with an
>infinite wiring space, infinitely long wires, and an infinite
>number of vias. To make everything work, there are restrictions on
>all of these as well as the differences in these from one wire to
>another.

Yes I know .. I spent 20 years working on / managing IC CAD software,
including routers, and I haven't quiet forgotten all of it yet. 8>.

You still haven't explained to my satisfaction why a =pin-out change=
can suddenly force the requirement for another two wiring planes. I can
see why an =additional= power requirement, or need for =additional=
signal conditioning (like you're running on the hairy edge, which is why
you needed buffered DIMMS in the first place) could up the wiring planes
needed, but a pinout change which magically needs two extra board layers
seems like something you could hardly devise if you worked at it.

--
GSV Three Minds in a Can
Outgoing Msgs are Turing Tested,and indistinguishable from human typing.
Anonymous
a b à CPUs
June 10, 2004 2:09:18 AM

Archived from groups: comp.sys.ibm.pc.hardware.chips (More info?)

GSV Three Minds in a Can wrote:

> Bitstring <3oadnSFRhc3F6FrdRVn-vA@adelphia.com>, from the
> wonderful person K Williams <krw@adelphia.net> said
>>GSV Three Minds in a Can wrote:
>>
>>> Bitstring <1086792271.200887@teapot.planet.gong>, from the
>>> wonderful person Rupert Pigott
>>> <roo@try-removing-this.darkboong.demon.co.uk> said <snip>
>>>>>>939 boards are also cheaper to make, 4 layer instead of 6.
>>>
>>>>> Why? I don't see the memory differences as significant.
>>>>
>>>>Same question crossed my mind. My guess is that they weren't
>>>>*too* bothered about Opteron boards being costly so they didn't
>>>>put much effort into keeping the layer count down. I'll bet that
>>>>some bright spark spotted an alternative pin-out that would
>>>>allow 4 layer boards sometime after S940 was released, hence the
>>>>magic new S939 that allows 4-Layer boards. :) 
>>>
>>> I'm not sure you can blame the pinout - I can't see why you
>>> can't route anything with 4 layers (actually, iirc, with 2 .. in
>>> theory, for constant width wires), unless/until you start
>>> needing massive power distribution and ground planes?
>>
>>Read that last sentence again. ;-)
>>
>>The reality is that you need the ground/power planes for
>>electrical reasons, other than the *MASSIVE* power these things
>>dissipate.
>>Think about 60W at ~1.25V. That's a wee bit of current. In
>>addition the impedance of the power distribution must be kept as
>>low as possible to reduce noise. The planes are also necessary to
>>keep the impedance of the signal lines constant. Two of the four
>>planes were power planes in the 486 days, without the massive
>>currents we now see.
>
>>Yes, you can theoretically wire anything using two layers, with an
>>infinite wiring space, infinitely long wires, and an infinite
>>number of vias. To make everything work, there are restrictions
>>on all of these as well as the differences in these from one wire
>>to another.
>
> Yes I know .. I spent 20 years working on / managing IC CAD
> software, including routers, and I haven't quiet forgotten all of
> it yet. 8>.
>
> You still haven't explained to my satisfaction why a =pin-out
> change= can suddenly force the requirement for another two wiring
> planes.

Oh, my... We're now asking the same question! My proposal is that
there is no pretense of a MP system with S939, so things are
relaxed a bit. That supposes that a Uni-S940 should be do-able
w/4-layers too. Maybe no one wants to design a S940 board for a
uni? I find that hard to swallow too. so...

> I can see why an =additional= power requirement, or need
> for =additional= signal conditioning (like you're running on the
> hairy edge, which is why you needed buffered DIMMS in the first
> place) could up the wiring planes needed, but a pinout change
> which magically needs two extra board layers seems like something
> you could hardly devise if you worked at it.

Registered DIMMs should need *less* care in layout, given the same
number. That's sorta the purpose behind registering.

--
Keith
Anonymous
a b à CPUs
June 10, 2004 2:10:39 AM

Archived from groups: comp.sys.ibm.pc.hardware.chips,comp.sys.intel (More info?)

Rupert Pigott wrote:

>> No. Socket 940, 939, and 754 were all announced at the same time.
>> Manufacturers were ready to start cranking out Socket 939 boards
>> more than a year ago but AMD just wasn't producing the chips.
>
>
> Eeek, I hadn't twigged 939 & 940 were launched at the same time. Seems
> a bit odd that the 4/6 layer difference exists in that case.

I saw my first Socket 939 board at the same time I saw my
first 940 board - in Feb 1993. At that time the AMD rep
at the demo was predicting general availability of socket
940 chips for April 2003 (which actually happened) and
socket 939 chips for Q3 2003. So far Q3 2004 is just a
few weeks away and no socket 939 chips seem to be in stock
in any stores.

(I also saw a 4-way board by NewiSys at that demo with
four 800 MHz Opties on it.:-) )
Anonymous
a b à CPUs
June 10, 2004 2:21:27 AM

Archived from groups: comp.sys.ibm.pc.hardware.chips (More info?)

K Williams wrote:

> Yousuf Khan wrote:
>
>
>>Rupert Pigott <roo@try-removing-this.darkboong.demon.co.uk> wrote:
>>
>>>Must admit I'm curious to see what kind of differences exist
>>>between 939 and 940.
>>
>>From a purely end-user perspective, the biggest difference between
>>939 and 940 is that 939 is for unbuffered DDR, while 940 is for
>>server-class buffered DDR. I assume that since 940 uses buffered
>>DDR, the potential exists to outfit 940 board with well over two
>>DIMMs per processor.
>
>
> But, but... Why would a 940 board that *only* supports four DIMMs
> be more complicated than a 939 board that only supports 4 DIMMs?
> DIMMs is Dimms (at least on this level).
>

There is nothing inherent in the Opty that limits motherboard
manufacturers to 4 DIMMs per processor. HP, for example, has
managed 8 in their 4-way systems.
Anonymous
a b à CPUs
June 10, 2004 10:11:33 AM

Archived from groups: comp.sys.ibm.pc.hardware.chips (More info?)

K Williams <krw@adelphia.net> wrote:
> Yousuf Khan wrote:
>> From a purely end-user perspective, the biggest difference between
>> 939 and 940 is that 939 is for unbuffered DDR, while 940 is for
>> server-class buffered DDR. I assume that since 940 uses buffered
>> DDR, the potential exists to outfit 940 board with well over two
>> DIMMs per processor.
>
> But, but... Why would a 940 board that *only* supports four DIMMs
> be more complicated than a 939 board that only supports 4 DIMMs?
> DIMMs is Dimms (at least on this level).

Probably because a 940 board will truly be able to support 4 DIMMs whereas a
939 board will more than likely be limited to two DIMMs in reality. When was
the last time you saw desktop boards routinely support more than two DIMMs?
I can only remember this happening back in the Pentium 1 days. Ever since
then, all I've ever seen is warnings against putting any more than two DIMMs
on any board even if it has space for more.

The only ones that can truly do it these days are the buffered DIMMs on
server boards.

Yousuf Khan
Anonymous
a b à CPUs
June 10, 2004 10:11:33 AM

Archived from groups: comp.sys.ibm.pc.hardware.chips (More info?)

K Williams <krw@adelphia.net> wrote:
>>>> Oh geez, now we got two Ed's, and they're gonna carry on a
>>>> confusing debate amongst themselves. :-)
>>>
>>> Nope. Same Ed, just a different moniker (look at the headings).
>>
>> If they are the same Ed, then why is one asking a question of the
>> other?
>
> My news server has a "Grumble" post threaded in between.

Yes, so does mine.

But the first Ed was saying that a Socket 745 required six layer mobos,
while Socket 939 requires only 4 layer. Then Grumble asked Ed #1 why that is
the case? Then Ed #2 responded to Grumble by answering with a
semi-rhetorical question: he asked if it wasn't true that S745 actually
requires 4 layers while S939 requires 6 layers, therefore contradicting Ed
#1.

>> This has the potential of making the Tony vs. Tony Hill
>> debates look comprehensible.
>
> ;-) That one got me!

I think if this sort of thing happens more often, then people who are using
things like OE-Quotefix to format their news reader styles, make sure that
they include at least the email addresses in the quote header.

Yousuf Khan
Anonymous
a b à CPUs
June 10, 2004 10:11:33 AM

Archived from groups: comp.sys.ibm.pc.hardware.chips (More info?)

K Williams <krw@adelphia.net> wrote:
> The only possible difference I can see is that the S939 boards won't
> be pretending to do more than one processor, so may be somewhat
> simpler that way. However that would assume the HT links aren't
> being used for anything else. So why all the pins?

Well as some people have pointed out in the case of Pentium 4's Socket
775 -- extra grounding.

Then again it's possible that maybe AMD is already preparing for DDR2, and
Socket 939 is a forward looking design preparing for that day.

Yousuf Khan
Anonymous
a b à CPUs
June 10, 2004 10:11:34 AM

Archived from groups: comp.sys.ibm.pc.hardware.chips,comp.sys.intel (More info?)

Rob Stow <rob.stow@sasktel.net> wrote:
>> Eeek, I hadn't twigged 939 & 940 were launched at the same time.
>> Seems a bit odd that the 4/6 layer difference exists in that case.
>
> I saw my first Socket 939 board at the same time I saw my
> first 940 board - in Feb 1993. At that time the AMD rep
> at the demo was predicting general availability of socket
> 940 chips for April 2003 (which actually happened) and
> socket 939 chips for Q3 2003. So far Q3 2004 is just a
> few weeks away and no socket 939 chips seem to be in stock
> in any stores.

Well, that's some forward thinking on AMD's part, they were already prepared
with a 64-bit processor socket ten years before they released it. :-)

Yousuf Khan
Anonymous
a b à CPUs
June 10, 2004 12:25:42 PM

Archived from groups: comp.sys.ibm.pc.hardware.chips,comp.sys.intel (More info?)

Yousuf Khan wrote:

> Rob Stow <rob.stow@sasktel.net> wrote:
>
>>>Eeek, I hadn't twigged 939 & 940 were launched at the same time.
>>>Seems a bit odd that the 4/6 layer difference exists in that case.
>>
>>I saw my first Socket 939 board at the same time I saw my
>>first 940 board - in Feb 1993. At that time the AMD rep
>>at the demo was predicting general availability of socket
>>940 chips for April 2003 (which actually happened) and
>>socket 939 chips for Q3 2003. So far Q3 2004 is just a
>>few weeks away and no socket 939 chips seem to be in stock
>>in any stores.
>
>
> Well, that's some forward thinking on AMD's part, they were already prepared
> with a 64-bit processor socket ten years before they released it. :-)
>
> Yousuf Khan
>
>

I meant last year. 2003, not 1993. 2004-1 = 1993 :-)
That has got to be the weirdest "typo" I've ever done.
Either that or my math qualifies me for a gov't job.
Anonymous
a b à CPUs
June 10, 2004 1:55:55 PM

Archived from groups: comp.sys.ibm.pc.hardware.chips (More info?)

Yousuf Khan wrote:

> But the first Ed was saying that a Socket 745 required six layer
> mobos, while Socket 939 requires only 4 layer.

Just a nit. Socket 754, not 745.

Go ahead, count 'em :-)

http://www.hardspell.com/newsimage/200392613445242585.j...
Anonymous
a b à CPUs
June 10, 2004 6:24:39 PM

Archived from groups: comp.sys.ibm.pc.hardware.chips (More info?)

Rob Stow wrote:

> K Williams wrote:
>
>> Yousuf Khan wrote:
>>
>>
>>>Rupert Pigott <roo@try-removing-this.darkboong.demon.co.uk>
>>>wrote:
>>>
>>>>Must admit I'm curious to see what kind of differences exist
>>>>between 939 and 940.
>>>
>>>From a purely end-user perspective, the biggest difference
>>>between 939 and 940 is that 939 is for unbuffered DDR, while 940
>>>is for server-class buffered DDR. I assume that since 940 uses
>>>buffered DDR, the potential exists to outfit 940 board with well
>>>over two DIMMs per processor.
>>
>>
>> But, but... Why would a 940 board that *only* supports four
>> DIMMs be more complicated than a 939 board that only supports 4
>> DIMMs? DIMMs is Dimms (at least on this level).
>>
>
> There is nothing inherent in the Opty that limits motherboard
> manufacturers to 4 DIMMs per processor. HP, for example, has
> managed 8 in their 4-way systems.

I didn't say there was. However many 940 boards only support four
(mine, fer instance;). Why would a 940 board that only supported
four take any more wiring channels than a 939 board that supported
four? For that matter, once you've fanned the signals out to the
memory modules, why would more memory modules require more layers?
Both assuming dual-channel, of course.

--
Keith
Anonymous
a b à CPUs
June 10, 2004 6:29:48 PM

Archived from groups: comp.sys.ibm.pc.hardware.chips (More info?)

Yousuf Khan wrote:

> K Williams <krw@adelphia.net> wrote:
>> Yousuf Khan wrote:
>>> From a purely end-user perspective, the biggest difference
>>> between 939 and 940 is that 939 is for unbuffered DDR, while 940
>>> is for server-class buffered DDR. I assume that since 940 uses
>>> buffered DDR, the potential exists to outfit 940 board with well
>>> over two DIMMs per processor.
>>
>> But, but... Why would a 940 board that *only* supports four
>> DIMMs be more complicated than a 939 board that only supports 4
>> DIMMs? DIMMs is Dimms (at least on this level).
>
> Probably because a 940 board will truly be able to support 4 DIMMs
> whereas a 939 board will more than likely be limited to two DIMMs
> in reality. When was the last time you saw desktop boards
> routinely support more than two DIMMs?

Lots of 'em support three. Remember, the 939 is still dual-channel
(right?). Are you suggesting that each channel will only have one
DIMM? Perhaps that's the only way to make it work, but I'm glad I
went with the 940, if so. Still once you've transformed the wires
out to the first DIMM, the traces to the second are pretty much
parallel, thus I still don't understand the issue.

> I can only remember this
> happening back in the Pentium 1 days. Ever since then, all I've
> ever seen is warnings against putting any more than two DIMMs on
> any board even if it has space for more.

Remember, It's dual channel, so there would only be two on each
channel (for a total of four).

> The only ones that can truly do it these days are the buffered
> DIMMs on server boards.

Sure at these speeds, and two per channel.

We're still missing something here.

--
Keith
Anonymous
a b à CPUs
June 10, 2004 6:32:43 PM

Archived from groups: comp.sys.ibm.pc.hardware.chips (More info?)

Yousuf Khan wrote:

> K Williams <krw@adelphia.net> wrote:
>> The only possible difference I can see is that the S939 boards
>> won't be pretending to do more than one processor, so may be
>> somewhat
>> simpler that way. However that would assume the HT links aren't
>> being used for anything else. So why all the pins?
>
> Well as some people have pointed out in the case of Pentium 4's
> Socket 775 -- extra grounding.

....and the 940 doesn't need 'em. I'm not buying that one either.

> Then again it's possible that maybe AMD is already preparing for
> DDR2, and Socket 939 is a forward looking design preparing for
> that day.

Perhaps. But why wouldn't they build that into the 940 footprint
too? Leave the desktop with faster memory (down the road, of
course) than the "server" chipset?

--
Keith
Anonymous
a b à CPUs
June 10, 2004 10:47:41 PM

Archived from groups: comp.sys.ibm.pc.hardware.chips (More info?)

K Williams <krw@adelphia.net> wrote:
> Yousuf Khan wrote:
>> Probably because a 940 board will truly be able to support 4 DIMMs
>> whereas a 939 board will more than likely be limited to two DIMMs
>> in reality. When was the last time you saw desktop boards
>> routinely support more than two DIMMs?
>
> Lots of 'em support three. Remember, the 939 is still dual-channel
> (right?). Are you suggesting that each channel will only have one
> DIMM? Perhaps that's the only way to make it work, but I'm glad I
> went with the 940, if so. Still once you've transformed the wires
> out to the first DIMM, the traces to the second are pretty much
> parallel, thus I still don't understand the issue.

Well, it's a dual-channel in the sense that the processor reads the lower
64-bits from the first DIMM and the upper 64-bits from the second DIMM. I
don't think it actually puts the two chips in separate banks and does bank
interleaving in the traditional dual-channel sense. This is how Pentium 4's
dual-channel works as well.

Yousuf Khan
Anonymous
a b à CPUs
June 10, 2004 10:47:41 PM

Archived from groups: comp.sys.ibm.pc.hardware.chips (More info?)

K Williams <krw@adelphia.net> wrote:
> Yousuf Khan wrote:
>> Well as some people have pointed out in the case of Pentium 4's
>> Socket 775 -- extra grounding.
>
> ...and the 940 doesn't need 'em. I'm not buying that one either.

Well, I don't either, just passing on what's been stated elsewhere.

>> Then again it's possible that maybe AMD is already preparing for
>> DDR2, and Socket 939 is a forward looking design preparing for
>> that day.
>
> Perhaps. But why wouldn't they build that into the 940 footprint
> too? Leave the desktop with faster memory (down the road, of
> course) than the "server" chipset?

That shouldn't be too difficult to accept. The server market is
traditionally more conservative about technological changes than the desktop
market is. So it's likely that once DDR2 has proven itself in the Athlon 64,
the Opteron will require a totally new socket other than S940.

Or perhaps Intel's FB-DIMM dreams will come true, and you won't be requiring
a new socket to move to DDR2, because it will work in existing DDR1 slots.

Yousuf Khan
Anonymous
a b à CPUs
June 10, 2004 10:47:42 PM

Archived from groups: comp.sys.ibm.pc.hardware.chips (More info?)

Yousuf Khan wrote:

> K Williams <krw@adelphia.net> wrote:
>> Yousuf Khan wrote:
>>> Probably because a 940 board will truly be able to support 4
>>> DIMMs whereas a 939 board will more than likely be limited to
>>> two DIMMs in reality. When was the last time you saw desktop
>>> boards routinely support more than two DIMMs?
>>
>> Lots of 'em support three. Remember, the 939 is still
>> dual-channel
>> (right?). Are you suggesting that each channel will only have
>> one
>> DIMM? Perhaps that's the only way to make it work, but I'm glad
>> I
>> went with the 940, if so. Still once you've transformed the
>> wires out to the first DIMM, the traces to the second are pretty
>> much parallel, thus I still don't understand the issue.
>
> Well, it's a dual-channel in the sense that the processor reads
> the lower 64-bits from the first DIMM and the upper 64-bits from
> the second DIMM. I don't think it actually puts the two chips in
> separate banks and does bank interleaving in the traditional
> dual-channel sense. This is how Pentium 4's dual-channel works as
> well.

How is this different than interleaving? An address and command is
sent to both channels and each fetch (in the burst) comes from
alternating channels. There are still two independent sets of
wires for the two channels, no? Otherwise I don't see what
dual-channel buys, electrically (note that I haven't looked that
closely).

--
Keith
Anonymous
a b à CPUs
June 11, 2004 3:38:41 AM

Archived from groups: comp.sys.ibm.pc.hardware.chips (More info?)

K Williams <krw@adelphia.net> wrote:
>> Well, it's a dual-channel in the sense that the processor reads
>> the lower 64-bits from the first DIMM and the upper 64-bits from
>> the second DIMM. I don't think it actually puts the two chips in
>> separate banks and does bank interleaving in the traditional
>> dual-channel sense. This is how Pentium 4's dual-channel works as
>> well.
>
> How is this different than interleaving? An address and command is
> sent to both channels and each fetch (in the burst) comes from
> alternating channels. There are still two independent sets of
> wires for the two channels, no? Otherwise I don't see what
> dual-channel buys, electrically (note that I haven't looked that
> closely).

Well, I guess it's just that in this case, both DIMMs are really just
extensions of one another inside the same bank. One row-column address is
asserted that activates both DIMMs. Whereas in the bank-interleaved
dual-channel, there has to two separate, but simultaneous, invocations of
rows and columns, one for each bank since they each DIMM resides in separate
banks.

Yousuf Khan
!