You say megabyte, I say mebibyte

Grinder wrote:
> kony wrote:
>> On Wed, 23 Jan 2008 20:32:39 GMT, Grinder
>> <grinder.TakeThisOut@no.spam.maam.com> wrote:
>>
>>> Ok, I officially regret my choice of title for this thread.
>>
>> But did Paul's link help? We were kinda wanting to hear if
>> you had resolved this.
>
> As to the first issue, about what technology is in use, we're at a bit
> of a stalemate because of conflicting descriptions on the web, and even
> from CPU-Z itself.
>
> As for the second, some general background, I think I've gained some
> ground.
>
>> If not, what are the markings on the individual chips on the
>> memory? IE - I'd assumed you gave the module label markings
>> already but not the chips themselves... thinking back it was
>> 16 on a double sided module IIRC.
>
> That's correct, there are 16 chips--eight on each side. The individual
> chips are marked as: 48LC16M8A2, which suggests 16Mx8 chips, 16 of them,
> for a total of 256MB/MiB.
>
> Searching 48LC16M8A2 at google turns up a number of hits that suggest a
> module made from these chips will be 128MB/MiB if it's single-sided, or
> 256MB/MiB if double-sided.
>
>> The main thing is, there's not likely anything you can do to
>> make this module show up as 256MB in that board if it's only
>> appearing as 128MB now. The CPU-Z report was very strange
>> as if the detected amount and SPD amount were backwards.
>
> Assuming all the speculation above is correct, how do I match it up with
> my mainboard's vocabulary? ie, Is it 64, 128 or 256 Mbit technology?

The 16M8 is a total of 128 megabits. Dividing the number by 8, that is a
16Megabyte memory. It has a x8 interface (so 8 of them are needed to
make a 64 bit wide array, forming one "rank" of memory).

So the question remains as to why the SPD information doesn't seem
to match. When using CPUZ, you select the slot to display from a tiny
menu. Does it still say the stick is only 128MB, when in fact it
physically contains 256MB of chips ?

In years past, before SPD existed, the BIOS determined the memory
size by "probing". We used to do that on some of the proprietary
computers we used to build. If you wrote past the end, during
the BIOS test, then the read back value would fail to verify.

The BIOS today may consult the SPD, for initial setup, but may still use
probing to determine the "end of memory" on each chip select. Otherwise,
for those computers where there is a density mismatch (440BX), how does the
computer actually set up the chip select for half the normal memory
space ? It doesn't seem to rely on the SPD alone. I would consider it
a bit strange though, if it actually searched past the SPD declared
value, because that would imply the SPD was really worthless
Well, at least it still has those timing values written in it.

By the way, the SPD on your stick of RAM is one of those "slightly
bogus" ones. The memory I have in this computer, has real serial
numbers written into the SPD chips. Some manufacturers prepare ROM
chips in advance, and each one of those is identical. They set the
serial number to FFFFFFFF. It is more trouble to use an EEPROM
for the SPD, and spend a few seconds writing a unique set of values
in the stick, but some follow the rules and do it right. The idea
is, it is to the manufacturer's advantage to put a unique number in
it, for easier tracking and verification during a warranty claim.
It would make it easier to identify a counterfeit, if the serial
numbers were unique and recorded at the plant.

Paul
 >> Stay informed about: You say megabyte, I say mebibyte 

Paul wrote:
> Grinder wrote:
>> kony wrote:
>>> On Wed, 23 Jan 2008 20:32:39 GMT, Grinder
>>> <grinder.DeleteThis@no.spam.maam.com> wrote:
>>>
>>>> Ok, I officially regret my choice of title for this thread.
>>>
>>> But did Paul's link help? We were kinda wanting to hear if
>>> you had resolved this.
>>
>> As to the first issue, about what technology is in use, we're at a bit
>> of a stalemate because of conflicting descriptions on the web, and
>> even from CPU-Z itself.
>>
>> As for the second, some general background, I think I've gained some
>> ground.
>>
>>> If not, what are the markings on the individual chips on the
>>> memory? IE - I'd assumed you gave the module label markings
>>> already but not the chips themselves... thinking back it was
>>> 16 on a double sided module IIRC.
>>
>> That's correct, there are 16 chips--eight on each side. The
>> individual chips are marked as: 48LC16M8A2, which suggests 16Mx8
>> chips, 16 of them, for a total of 256MB/MiB.
>>
>> Searching 48LC16M8A2 at google turns up a number of hits that suggest
>> a module made from these chips will be 128MB/MiB if it's single-sided,
>> or 256MB/MiB if double-sided.
>>
>>> The main thing is, there's not likely anything you can do to
>>> make this module show up as 256MB in that board if it's only
>>> appearing as 128MB now. The CPU-Z report was very strange
>>> as if the detected amount and SPD amount were backwards.
>>
>> Assuming all the speculation above is correct, how do I match it up
>> with my mainboard's vocabulary? ie, Is it 64, 128 or 256 Mbit
>> technology?
>
> The 16M8 is a total of 128 megabits. Dividing the number by 8, that is a
> 16Megabyte memory. It has a x8 interface (so 8 of them are needed to
> make a 64 bit wide array, forming one "rank" of memory).
>
> So the question remains as to why the SPD information doesn't seem
> to match. When using CPUZ, you select the slot to display from a tiny
> menu. Does it still say the stick is only 128MB, when in fact it
> physically contains 256MB of chips ?

Absolutely.

I installed the module back into the computer that can only see it as
128MB, and ran CPU-Z. The SPD tab was identical, unsurprisingly, the
system tab was a bit different:

1st Machine 2nd Machine
GENERAL FRAME
Type: SDRAM SDRAM
Size: 256 MBytes 128 MBytes
Bank Interleave: none (blank)

TIMINGS FRAME
Frequency: 100.0 MHz 132.9 MHz
FSB:DRAM: (blank) 3:4
tCL: 3.0 clocks 3.0 clocks
tRCD: 3 clocks 3 clocks
tRP: 3 clocks 3 clocks
tRAS: 6 clocks 6 clocks

> In years past, before SPD existed, the BIOS determined the memory
> size by "probing". We used to do that on some of the proprietary
> computers we used to build. If you wrote past the end, during
> the BIOS test, then the read back value would fail to verify.
>
> The BIOS today may consult the SPD, for initial setup, but may still use
> probing to determine the "end of memory" on each chip select. Otherwise,
> for those computers where there is a density mismatch (440BX), how does the
> computer actually set up the chip select for half the normal memory
> space ? It doesn't seem to rely on the SPD alone. I would consider it
> a bit strange though, if it actually searched past the SPD declared
> value, because that would imply the SPD was really worthless
> Well, at least it still has those timing values written in it.
>
> By the way, the SPD on your stick of RAM is one of those "slightly
> bogus" ones. The memory I have in this computer, has real serial
> numbers written into the SPD chips. Some manufacturers prepare ROM
> chips in advance, and each one of those is identical. They set the
> serial number to FFFFFFFF. It is more trouble to use an EEPROM
> for the SPD, and spend a few seconds writing a unique set of values
> in the stick, but some follow the rules and do it right. The idea
> is, it is to the manufacturer's advantage to put a unique number in
> it, for easier tracking and verification during a warranty claim.
> It would make it easier to identify a counterfeit, if the serial
> numbers were unique and recorded at the plant.
>
> Paul
 >> Stay informed about: You say megabyte, I say mebibyte 

Grinder wrote:
>
> Absolutely.
>
> I installed the module back into the computer that can only see it as
> 128MB, and ran CPU-Z. The SPD tab was identical, unsurprisingly, the
> system tab was a bit different:
>
> 1st Machine 2nd Machine
> GENERAL FRAME
> Type: SDRAM SDRAM
> Size: 256 MBytes 128 MBytes
> Bank Interleave: none (blank)
>
> TIMINGS FRAME
> Frequency: 100.0 MHz 132.9 MHz
> FSB:DRAM: (blank) 3:4
> tCL: 3.0 clocks 3.0 clocks
> tRCD: 3 clocks 3 clocks
> tRP: 3 clocks 3 clocks
> tRAS: 6 clocks 6 clocks
>

In CPUZ, if you go to "About", there is a Register Dump. It will
dump a file "cpuz.txt". In there, you'll see something like

Dump Module #1
0 1 2 3 4 5 6 7 8 9 A B C D E F
00 80 08 07 0D 0A 02 40 00 04 50 60 00 82 08 00 01
10 0E 04 04 01 02 20 C0 00 00 00 00 28 28 28 28 40
20 70 70 40 40 00 00 00 00 00 37 46 30 28 50 00 00
30 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 85 ...

Well, you won't see those particular values, because that is
a dump of my 512MB DDR stick, while yours is SDRAM.

JEDEC has documents to decode those numbers. This doc is
just a guess, as being the one for SDRAM. I used an outside
search engine, rather than messing with the JEDEC search option.
Byte 5 (decimal) and Byte 31 (decimal), as seen on page 20,
define the capacity of each bank. Byte 5 should say 0x02 hex,
meaning there are two banks (double sided). Byte 31 should
read 0x20 hex, to indicate 128MB per bank, for a total of
256MB.

http://www.jedec.org/download/search/4_01_02_05R12.PDF

Have a look at the dump table, and see what your byte 5
and byte 31 show for the module. (Note - the above table
is oriented in hex form, so byte 31 decimal is the second row
right-most byte, as in byte 0x1F, and has a value of 0x40.)

The stuff above 3F, I didn't copy, because it has things like
the serial number and manufacturer ID.

Paul
 >> Stay informed about: You say megabyte, I say mebibyte 

Paul wrote:
> Grinder wrote:
>>
>> Absolutely.
>>
>> I installed the module back into the computer that can only see it as
>> 128MB, and ran CPU-Z. The SPD tab was identical, unsurprisingly, the
>> system tab was a bit different:
>>
>> 1st Machine 2nd Machine
>> GENERAL FRAME
>> Type: SDRAM SDRAM
>> Size: 256 MBytes 128 MBytes
>> Bank Interleave: none (blank)
>>
>> TIMINGS FRAME
>> Frequency: 100.0 MHz 132.9 MHz
>> FSB:DRAM: (blank) 3:4
>> tCL: 3.0 clocks 3.0 clocks
>> tRCD: 3 clocks 3 clocks
>> tRP: 3 clocks 3 clocks
>> tRAS: 6 clocks 6 clocks
>>
>
> In CPUZ, if you go to "About", there is a Register Dump. It will
> dump a file "cpuz.txt". In there, you'll see something like
>
> Dump Module #1
> 0 1 2 3 4 5 6 7 8 9 A B C D E F
> 00 80 08 07 0D 0A 02 40 00 04 50 60 00 82 08 00 01
> 10 0E 04 04 01 02 20 C0 00 00 00 00 28 28 28 28 40
> 20 70 70 40 40 00 00 00 00 00 37 46 30 28 50 00 00
> 30 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 85 ...
>
> Well, you won't see those particular values, because that is
> a dump of my 512MB DDR stick, while yours is SDRAM.
>
> JEDEC has documents to decode those numbers. This doc is
> just a guess, as being the one for SDRAM. I used an outside
> search engine, rather than messing with the JEDEC search option.
> Byte 5 (decimal) and Byte 31 (decimal), as seen on page 20,
> define the capacity of each bank. Byte 5 should say 0x02 hex,
> meaning there are two banks (double sided). Byte 31 should
> read 0x20 hex, to indicate 128MB per bank, for a total of
> 256MB.
>
> http://www.jedec.org/download/search/4_01_02_05R12.PDF
>
> Have a look at the dump table, and see what your byte 5
> and byte 31 show for the module. (Note - the above table
> is oriented in hex form, so byte 31 decimal is the second row
> right-most byte, as in byte 0x1F, and has a value of 0x40.)
>
> The stuff above 3F, I didn't copy, because it has things like
> the serial number and manufacturer ID.

This is what I get:

Dump Module #1
0 1 2 3 4 5 6 7 8 9 A B C D E F
00 80 08 04 0C 0A 01 40 00 01 75 54 00 80 08 00 01
10 8F 04 04 01 01 00 0E 00 00 00 00 14 0F 14 2D 20
20 15 08 15 80 00 00 00 00 00 00 00 00 00 00 00 00
30 00 00 00 00 00 00 00 00 00 00 00 00 00 00 12 AD
...

I guess maybe there's the rub--byte 5 is 1, or single-sided.
 >> Stay informed about: You say megabyte, I say mebibyte 

"kony" <spam.RemoveThis@spam.com> wrote in message
news:daffp3leu3jq1mr1c7iqq1mmfe6hhckmfk@4ax.com...
> On Tue, 22 Jan 2008 22:33:50 -0000, "GT"
> <ContactGT_remove_.RemoveThis@hotmail.com> wrote:
>
>>we are just never going to agree on this one!
>
> Ok, then let's not spend any more time on it.

Agreed! More useful things to do with our time!
 >> Stay informed about: You say megabyte, I say mebibyte 

"Grinder" <grinder.RemoveThis@no.spam.maam.com> wrote in message
news:HvNlj.313269$Fc.231252@attbi_s21...
> Ok, I officially regret my choice of title for this thread.

My fault!
 >> Stay informed about: You say megabyte, I say mebibyte 

On 4 Feb, 14:35, "GT" <ContactGT_remo....DeleteThis@hotmail.com> wrote:
<snip>
>
> In short, this argument will run and run for years and never be concluded

I will not assume that you are illogical. If you were, then this would
indeed never be concluded.



> because certain companies and individuals refuse to use the clearly
> specified industry standard terms/abbreviations correctly. The term MB means
> MegaBytes. Mega is 10^6. Mega never means 2^20. That is Mebi.
>

The term " Mebi ", and the rest are fairly new terms, introduced by a
standards authority, for obvious reasons. And In the hope that they
will become mainstream.

The term Megabyte, to the people that used it - programmers - when
dealing with RAM, referred to - exactly - 2^20.

(likewise the rest obviously, as you know. kilo, mega, blah blah blah.
I am just mentioning mega alone usually - for very beneficial brevity)

That covers our possible points of current disagreement..

So my position on SI notation, I don`t know if this tallies with
yours.

Of course, originally, the terms Mega, Giga, e.t.c. existed before
anybody used the word Byte. I guess they existed to make the
mathematical "scientific notation" easier to verbalise and think
about(in words). And so Mega, Giga, Tera , Nano, Milli, and the rest,
only ever meant 10^x where x is one of a range of positive or negative
numbers that is a multiple of 3. (oh, and centi
i.e. 10^-2 seems to be there too!).


Given the history. First the SI notation, then the computer scientists
using Megabyte in the context of RAM, to mean 2^20.

And then the very belated and unsuccessful attempt by a standards
authority to try to "fix the problem". Programmers working with RAM,
will not change a wonderful tradition of saying Megabyte and meaning
2^20. MebiByte is a bit of a tongue twister, like Megabyte gone wrong.
It`s like Peter Piper Picked.. In order to say it you have to shout
out the consonants. Infact, if you got Peter Piper picking MebiBytes,
you`d really keep kids occupied.

Given that - the history anyway - I would not say as you do that
Megabyte means 10^6 bytes and only 10^6 bytes.

Infact, I remember hearing of Mebibyte in an article in scott
mueller`s fat book.. Then it became some odd news article that people
scoffed at, and I recall lots of people saying they refused to use
those words. Standards Authorities are there to help. They cannot
trounce on an established tradition and expect it to work out. It has
not worked out for them. People/Society make words.. As long as
those people are techies, and not marketters, I do not have a problem
with it. I think it`s good. Techies when they used the term, knew what
they meant.. And so do techies now. And there is a logical reason why
they would use one or the other. So these terms like Mebi and Kili and
whatever, have not gained such widespread acceptance.. Certainly not
enough for you to take the line you are taking

It may be a bit of a nuisance for an end user trying to figure out how
much hard drive space he has - not out of technical interest - they do
not know binary either. They simply want to cry over a few gigabytes,
to make a fuss. Screw them. They are alien retards, and it`s not
their game.
And if they want to know, fine. HDD manufacturers label it as Mega
being 10^x. Smaller unit. Bigger number of them. Confusion cleared up.
As has been mentioned many times already. If they cannot use Google,
then screw them again. They can pay for you to come round, and they
do.

Assembly language programmers did not get confused either, talking of
Kilobytes of RAM, and meaning 2^10 not 10^3.. (and if where there
are still assembly language programmers, I am sure they still do not
get confused)

<snip>
 >> Stay informed about: You say megabyte, I say mebibyte 

[series of strangely linked quotes all snipper]
>
> I don`t plan on reading his badly presented points again, but I think
> he was saying regarding hard drive capacities, that windows, by
> reporting the 2^x number, reports it wrongly..
> And so hard drive marketters, by using the 10^x format, are correct.
> (which also gives them a higher number btw, since it is a smaller unit
> than the corresponding 2^y. One reason why many consider them to be
> doing so for dishonest reasons).

I'm sorry, I couldn't follow your last post - too bitsy!

In short, this argument will run and run for years and never be concluded
because certain companies and individuals refuse to use the clearly
specified industry standard terms/abbreviations correctly. The term MB means
MegaBytes. Mega is 10^6. Mega never means 2^20. That is Mebi.

Hard disk manufacturers *correctly* advertise their hard drives in MegaBytes
(millions of bytes). Windows reports a size in MebiBytes, but labels it
*incorrectly* as MegaBytes (MB). The reason Windows is incorrect is that the
number it displays is not MegaBytes (MB), but actually MebiBytes(MiB). There
is no problem with the 'binary' calculations using 2^10, 2^20 etc, the
problem is simply that when we use these numbers, we can't label the answer
as MB.

Here is a page on the National Institute of Standards and Technology website
that explains the terms clearly and simply...

http://physics.nist.gov/cuu/Units/binary.html
 >> Stay informed about: You say megabyte, I say mebibyte 

On Mon, 4 Feb 2008 14:35:39 -0000, "GT"
<ContactGT_remove_ RemoveThis @hotmail.com> wrote:


>I'm sorry, I couldn't follow your last post - too bitsy!
>
>In short, this argument will run and run for years and never be concluded
>because certain companies and individuals refuse to use the clearly
>specified industry standard terms/abbreviations correctly.

>The term MB means
>MegaBytes. Mega is 10^6.

Except in the computer industry, which already had defined
that term, probably long before you were born.

>Mega never means 2^20. That is Mebi.

Except in the computer industry. Their use was standard and
far predates 1998. Quite simply if NIST wanted to create a
new term it would have to apply to the new value not the
binary one because in this industry the binary one is taken,
reserved, unavailable for them to *assign* some value to.

That's what a standard is, the thing that came first and
that everyone used. It doesn't change because someone else
says so, dozens of years later.
 >> Stay informed about: You say megabyte, I say mebibyte 

"kony" <spam.TakeThisOut@spam.com> wrote in message
news:24geq3pdd50a1p5q4ip3epvhhdohsi829c@4ax.com...
> On Mon, 4 Feb 2008 14:35:39 -0000, "GT"
> <ContactGT_remove_.TakeThisOut@hotmail.com> wrote:
>
>
>>I'm sorry, I couldn't follow your last post - too bitsy!
>>
>>In short, this argument will run and run for years and never be concluded
>>because certain companies and individuals refuse to use the clearly
>>specified industry standard terms/abbreviations correctly.
>
>>The term MB means
>>MegaBytes. Mega is 10^6.
>
> Except in the computer industry, which already had defined
> that term, probably long before you were born.

You are wrong. The computer industry didn't define Mega. The term Mega was
clearly defined to mean 10^6 before you, I and even Babbage was born, so the
computer industry didn't define it, they just misused it!
 >> Stay informed about: You say megabyte, I say mebibyte 

On Tue, 5 Feb 2008 09:48:16 -0000, "GT"
<ContactGT_remove_.DeleteThis@hotmail.com> wrote:

>"kony" <spam.DeleteThis@spam.com> wrote in message
>news:24geq3pdd50a1p5q4ip3epvhhdohsi829c@4ax.com...
>> On Mon, 4 Feb 2008 14:35:39 -0000, "GT"
>> <ContactGT_remove_.DeleteThis@hotmail.com> wrote:
>>
>>
>>>I'm sorry, I couldn't follow your last post - too bitsy!
>>>
>>>In short, this argument will run and run for years and never be concluded
>>>because certain companies and individuals refuse to use the clearly
>>>specified industry standard terms/abbreviations correctly.
>>
>>>The term MB means
>>>MegaBytes. Mega is 10^6.
>>
>> Except in the computer industry, which already had defined
>> that term, probably long before you were born.
>
>You are wrong. The computer industry didn't define Mega. The term Mega was
>clearly defined to mean 10^6 before you, I and even Babbage was born, so the
>computer industry didn't define it, they just misused it!

The industry defined the value of the terms megabyte, etc.
Only AFTER they had done so and used the terms as standards
for dozens of years did people like you come along and thing
the _STANDARD_ should change. That's the opposite of what a
standard is.

I agree the computer industry should not have taken the
terms mega, etc, because it is a misuse of them to have done
so in the first place- but they _did_ take these terms and
did use them to mean a specific value unto the point that it
_is_ a standard in the industry for many many years.

If you or NIST wants to intrduce a new term for a new value,
go right ahead and do so, but not trying to revalue the
existing standardized term.
 >> Stay informed about: You say megabyte, I say mebibyte 

>>>>I'm sorry, I couldn't follow your last post - too bitsy!
>>>>
>>>>In short, this argument will run and run for years and never be
>>>>concluded
>>>>because certain companies and individuals refuse to use the clearly
>>>>specified industry standard terms/abbreviations correctly.
>>>
>>>>The term MB means
>>>>MegaBytes. Mega is 10^6.
>>>
>>> Except in the computer industry, which already had defined
>>> that term, probably long before you were born.
>>
>>You are wrong. The computer industry didn't define Mega. The term Mega was
>>clearly defined to mean 10^6 before you, I and even Babbage was born, so
>>the
>>computer industry didn't define it, they just misused it!
>
> The industry defined the value of the terms megabyte, etc.
> Only AFTER they had done so and used the terms as standards
> for dozens of years did people like you come along and thing
> the _STANDARD_ should change. That's the opposite of what a
> standard is.

No, the _STANDARD_ was defined centuries ago. The computer industry simply
misused it for their own purposes. The term Mega was never adopted as a
standard for 2^20. Sure, the industry used it for ages, but it has never
been the _STANDARD_ interpretation of Mega.

> I agree the computer industry should not have taken the
> terms mega, etc, because it is a misuse of them to have done
> so in the first place- but they _did_ take these terms and
> did use them to mean a specific value unto the point that it
> _is_ a standard in the industry for many many years.

Not it is not a standard - it is a misuse of the standard term and has been
for years.

> If you or NIST wants to intrduce a new term for a new value,
> go right ahead and do so, but not trying to revalue the
> existing standardized term.

The term MiB for the value 2^20 has taken years to achieve recognition, but
it is now a clearly defined and the interim solution of misusing MB is now
superceeded.
 >> Stay informed about: You say megabyte, I say mebibyte 

On Wed, 6 Feb 2008 09:47:22 -0000, "GT"
<ContactGT_remove_ RemoveThis @hotmail.com> wrote:


>> The industry defined the value of the terms megabyte, etc.
>> Only AFTER they had done so and used the terms as standards
>> for dozens of years did people like you come along and thing
>> the _STANDARD_ should change. That's the opposite of what a
>> standard is.
>
>No, the _STANDARD_ was defined centuries ago.

Nonsense, centuries ago there was not any industry using
"gigabyte". There was only ONE widespread use of the term,
in the industry that DEFINED it.


>The computer industry simply
>misused it for their own purposes.

We can fairly say that when they adopted the term, their
definition was unconventional, but by the vast billions and
billions of people that then used it, it became a standard.

You don't really understand what it takes to become a
standard do you? It does not take waiting until NIST, or
you, decide to voice an opinion.

>The term Mega was never adopted as a
>standard for 2^20. Sure, the industry used it for ages, but it has never
>been the _STANDARD_ interpretation of Mega.

WRONG.

It was adopted, and did become the standard.
Dozens of years ago. This is incredibly obvious, so much
so that you feel you need to argue against it.
 >> Stay informed about: You say megabyte, I say mebibyte