A computer components & hardware forum. HardwareBanter

If this is your first visit, be sure to check out the FAQ by clicking the link above. You may have to register before you can post: click the register link above to proceed. To start viewing messages, select the forum that you want to visit from the selection below.

Go Back   Home » HardwareBanter forum » General Hardware & Peripherals » Homebuilt PC's
Site Map Home Register Authors List Search Today's Posts Mark Forums Read Web Partners

wasting the memory speed



 
 
Thread Tools Display Modes
  #1  
Old April 10th 04, 08:03 PM
esara
external usenet poster
 
Posts: n/a
Default wasting the memory speed

Accroding to this diagaram
http://www.viatech.com/en/k7-series/kt600.jsp

If for example My computer has DDR 400 and the CPU FSB is 400. In this
case I am wasting my memory speed. I mean the memory can send/receive
data to/from the chipset at speed 800Mhz(400*2) but the CPU which
suppose to handle these data can only work at FSB =400Mhz (I know the
internal speed is more than 400Mhz) but what I want to say is that
while the memory can supply data to the CPU at speed 800 the FSB of
the CPU can only take data at speed 400Mhz (so the FSB will slow down
the traffic). In this case this is not good design, the better is to
have CPU with FSB 800 or Why I bother to install memory DDR400, while
DDR200 will give the same performance provided that the CPU is FSB400.

Am I right?? any help would be very much apprciate it. Thanks.


Thanks
  #2  
Old April 10th 04, 09:01 PM
Jim
external usenet poster
 
Posts: n/a
Default

Ok, one problem here is that the marketing hype leads the advertising people
to not quite state things correctly. Let's forget the details of that ad,
and talk as accurately as possible.

The CPU FSB is more correctly stated as 200MHz (actual), but since it
employs DDR technology, it's *effective* rate is 400MHz. The "400MHz"
number you see in the ad is actually misleading, the CPU FSB does NOT
physically *run* at 400MHz, it runs at 200MHz (that's what would show up on
a scope!), *but*, because the DDR technology it employs allows data to
travel BOTH on the up and down side of each cycle (per MHz), it's *behaving*
as if it data was traveling on ONE side of the cycle, but at 400MHz! Get
it? It's a marketing gimic, the CPU isn't actually *faster*, it's more
*efficient* (2x in fact) at the same speed of a 200MHz processor (1x) that
does NOT employ DDR technology. But the marketing guys don't like that sort
of explanation, they'd rather claim the CPU is a 400MHz processor! Looks
better in the advertising than "this is a 200Mhz processor that's more
efficient and acts like a 400MHz processor". Not nearly as appealing,
right?

Same thing w/ Intel, they say their top-end processor is an 800Mhz CPU.
Technically, WRONG! It's a 200MHz processor that uses DDR technology (2x)
*and* twice as wide a data path (2 x 8 bit), making for an *effective* rate
of 200MHz x 4 = 800MHz. Whalla, the marketing guys are happy again.

Now to the memory. DDR400 (or PC3200) memory is marketed in exactly the
same fashion. It's 200MHz memory, but due to DDR technology, its
*effective* speed is 400MHz. But just as w/ the CPU FSB, it's really only
running at 200MHz (on the scope). Again, it's marketing hype. If we want
to be absolutely precise, that DDR400 (PC3200) memory is 200MHz x 2 (effect
of DDR) x 8 (bits wide) = PC3200. Some advertising people like the sound of
PC3200 instead of DDR400, so sometimes you see PC3200 referenced instead
(3200 sound more impressive than 400, I suppose). The PC3200 label also
implies it's effective bandwidth (3.2Gb/sec).

Bottomline, that AMD "400" CPU and DDR400 memory are, in fact, in PERFECT
sync, both are running 200MHz *actual*, on the "scope". What makes them
different/better is they are more efficient at that 200MHz speed than other
200MHz components that do NOT employ technologies like DDR.

Everything got all complicated in this area once DDR technology came into
the scene. Unfortunately, you can't take the advertising statements too
literally anymore. AMD did the same thing w/ their processors. Is an AMD
Athlon 2600+ actually running at 2.6GHz? No way, that's the hype, it's
actually 2.08GHz "on scope", AMD is merely claiming 2600 (2.6GHz) is the
*effective* performance compared to a 2.6GHz Intel CPU. Fact or fiction?
You decide.

HTH

Jim




"esara" wrote in message
om...
Accroding to this diagaram
http://www.viatech.com/en/k7-series/kt600.jsp

If for example My computer has DDR 400 and the CPU FSB is 400. In this
case I am wasting my memory speed. I mean the memory can send/receive
data to/from the chipset at speed 800Mhz(400*2) but the CPU which
suppose to handle these data can only work at FSB =400Mhz (I know the
internal speed is more than 400Mhz) but what I want to say is that
while the memory can supply data to the CPU at speed 800 the FSB of
the CPU can only take data at speed 400Mhz (so the FSB will slow down
the traffic). In this case this is not good design, the better is to
have CPU with FSB 800 or Why I bother to install memory DDR400, while
DDR200 will give the same performance provided that the CPU is FSB400.

Am I right?? any help would be very much apprciate it. Thanks.


Thanks



  #3  
Old April 10th 04, 09:40 PM
Last Boy Scout
external usenet poster
 
Posts: n/a
Default

On 10 Apr 2004 12:03:21 -0700, (esara) wrote:

I will give you some examples. When you use DMA and write files or
data, it does not go through the processor. The processor uses
message codes to know when it is done writing or whatever. Dont
assume everything is moving through the CPU. The processor also has
multiple execution pipelines. These pipelines all go at once. A CPU
is not just one pipeline, it is a lot more complicated at once. Also
some data is in a cache that is used over and over.

Accroding to this diagaram
http://www.viatech.com/en/k7-series/kt600.jsp

If for example My computer has DDR 400 and the CPU FSB is 400. In this
case I am wasting my memory speed. I mean the memory can send/receive
data to/from the chipset at speed 800Mhz(400*2) but the CPU which
suppose to handle these data can only work at FSB =400Mhz (I know the
internal speed is more than 400Mhz) but what I want to say is that
while the memory can supply data to the CPU at speed 800 the FSB of
the CPU can only take data at speed 400Mhz (so the FSB will slow down
the traffic). In this case this is not good design, the better is to
have CPU with FSB 800 or Why I bother to install memory DDR400, while
DDR200 will give the same performance provided that the CPU is FSB400.

Am I right?? any help would be very much apprciate it. Thanks.


Thanks


  #4  
Old April 10th 04, 09:56 PM
sooky grumper
external usenet poster
 
Posts: n/a
Default

Sorry for top posting, but if you didn't cut and paste that Jim you
certainly should post it to a website. Great explaination.

Jim wrote:
Ok, one problem here is that the marketing hype leads the advertising people
to not quite state things correctly. Let's forget the details of that ad,
and talk as accurately as possible.

The CPU FSB is more correctly stated as 200MHz (actual), but since it
employs DDR technology, it's *effective* rate is 400MHz. The "400MHz"
number you see in the ad is actually misleading, the CPU FSB does NOT
physically *run* at 400MHz, it runs at 200MHz (that's what would show up on
a scope!), *but*, because the DDR technology it employs allows data to
travel BOTH on the up and down side of each cycle (per MHz), it's *behaving*
as if it data was traveling on ONE side of the cycle, but at 400MHz! Get
it? It's a marketing gimic, the CPU isn't actually *faster*, it's more
*efficient* (2x in fact) at the same speed of a 200MHz processor (1x) that
does NOT employ DDR technology. But the marketing guys don't like that sort
of explanation, they'd rather claim the CPU is a 400MHz processor! Looks
better in the advertising than "this is a 200Mhz processor that's more
efficient and acts like a 400MHz processor". Not nearly as appealing,
right?

Same thing w/ Intel, they say their top-end processor is an 800Mhz CPU.
Technically, WRONG! It's a 200MHz processor that uses DDR technology (2x)
*and* twice as wide a data path (2 x 8 bit), making for an *effective* rate
of 200MHz x 4 = 800MHz. Whalla, the marketing guys are happy again.

Now to the memory. DDR400 (or PC3200) memory is marketed in exactly the
same fashion. It's 200MHz memory, but due to DDR technology, its
*effective* speed is 400MHz. But just as w/ the CPU FSB, it's really only
running at 200MHz (on the scope). Again, it's marketing hype. If we want
to be absolutely precise, that DDR400 (PC3200) memory is 200MHz x 2 (effect
of DDR) x 8 (bits wide) = PC3200. Some advertising people like the sound of
PC3200 instead of DDR400, so sometimes you see PC3200 referenced instead
(3200 sound more impressive than 400, I suppose). The PC3200 label also
implies it's effective bandwidth (3.2Gb/sec).

Bottomline, that AMD "400" CPU and DDR400 memory are, in fact, in PERFECT
sync, both are running 200MHz *actual*, on the "scope". What makes them
different/better is they are more efficient at that 200MHz speed than other
200MHz components that do NOT employ technologies like DDR.

Everything got all complicated in this area once DDR technology came into
the scene. Unfortunately, you can't take the advertising statements too
literally anymore. AMD did the same thing w/ their processors. Is an AMD
Athlon 2600+ actually running at 2.6GHz? No way, that's the hype, it's
actually 2.08GHz "on scope", AMD is merely claiming 2600 (2.6GHz) is the
*effective* performance compared to a 2.6GHz Intel CPU. Fact or fiction?
You decide.

HTH

Jim




"esara" wrote in message
om...

Accroding to this diagaram
http://www.viatech.com/en/k7-series/kt600.jsp

If for example My computer has DDR 400 and the CPU FSB is 400. In this
case I am wasting my memory speed. I mean the memory can send/receive
data to/from the chipset at speed 800Mhz(400*2) but the CPU which
suppose to handle these data can only work at FSB =400Mhz (I know the
internal speed is more than 400Mhz) but what I want to say is that
while the memory can supply data to the CPU at speed 800 the FSB of
the CPU can only take data at speed 400Mhz (so the FSB will slow down
the traffic). In this case this is not good design, the better is to
have CPU with FSB 800 or Why I bother to install memory DDR400, while
DDR200 will give the same performance provided that the CPU is FSB400.

Am I right?? any help would be very much apprciate it. Thanks.


Thanks






--
spammage trappage: replace fishies_ with yahoo
  #5  
Old April 12th 04, 11:35 PM
external usenet poster
 
Posts: n/a
Default

On Sat, 10 Apr 2004 13:01:39 -0700, "Jim" wrote:

Ok, one problem here is that the marketing hype leads the advertising people
to not quite state things correctly. Let's forget the details of that ad,
and talk as accurately as possible.

The CPU FSB is more correctly stated as 200MHz (actual), but since it
employs DDR technology, it's *effective* rate is 400MHz. The "400MHz"
number you see in the ad is actually misleading, the CPU FSB does NOT
physically *run* at 400MHz, it runs at 200MHz (that's what would show up on
a scope!), *but*, because the DDR technology it employs allows data to
travel BOTH on the up and down side of each cycle (per MHz), it's *behaving*
as if it data was traveling on ONE side of the cycle, but at 400MHz! Get
it? It's a marketing gimic, the CPU isn't actually *faster*, it's more
*efficient* (2x in fact) at the same speed of a 200MHz processor (1x) that
does NOT employ DDR technology.


The clock is just that. It's just a clock. It might run at 'just' 200
MHz, but it's _NOT_ the speed of the bus! The speed of the bus is
400MHz, and there's no marketing gimmick about that. That is indeed
the _speed_!
Now what purpose does the clock have?
It's like this: When data is transmitted on the FSB, data is
represented on the 72 pins (64 + ecc) out by voltage levels.
But as the data have to change to a new value, how do we know _when_
to read the data? When are all the leads ready, and the data correct?
That's where the clock comes in. The clock syncs the transfer. And
that is the _ONLY_ purpose of any clock. The clock tells _when_ the
data is supposed to be ready, and ok to read.
Now, you can sync on the clocks rising flank, or you can sync on the
falling flank. ... - Or, _both_!
AMD's FSB is the DEC Alpha EV6 protocol bus. And this happens to sync
on both rising and falling flanks.

So for a bus speed of 400MHz, - a 200MHz clock is _required_!
The data on the pins change 400 million times per second. That's the
_SPEED_ of bus! And that's only thing that matters, and marketing is
entirely 100% correct in stating that FSB speed is 400MHz.
That is not marketing hype or a gimmick.

I suppose I have to blame Intel marketing for everybody to be so damn
hung up on clockrates. - Hey, guys, - it's just a clock!

As for DDR ram, I don't know how it works, but I assume the actual
transfer is something similar to the EV6 bus. But there are more
complex things involved with memory access.
DDR speed only affects the bandwidth that memoryblocks can be
transferred with. An actual access is slower. A long chain of things
need to respond,

AMD did the same thing w/ their processors. Is an AMD
Athlon 2600+ actually running at 2.6GHz? No way, that's the hype, it's
actually 2.08GHz "on scope", AMD is merely claiming 2600 (2.6GHz) is the
*effective* performance compared to a 2.6GHz Intel CPU. Fact or fiction?
You decide.


It's not quite the same thing. But it's the same in the sense that the
clockrate is not the "speed" of any cpu. A clock is still just a
clock. The clock doesn't do the actual work.
Since Intel went for misleading the market with the P4, -

(The P4 is unique, since it's _less_ efficient per clockcycle than
preceding cpus, Pentium, PII, PIII. All other new generation cpus,
have always been more efficient per clockcycle than previous. That's a
natural outcome of an increased number of transistors. In the case of
the P4 and Prescott, their increased transistor counts are dedicated
to make it possible to run a higher clock. Not to do more work. This
is not good engineering. The P4 is engineered, intentionally, to run
at as high clock as possible. Certainly sacrificing performance! And
_THIS_ is indeed entirely marketing hype. A "gimic".)

, - AMD went for a 'rating' naming system instead. ...And so will
Intel shortly...

ancra

  #6  
Old April 12th 04, 11:36 PM
external usenet poster
 
Posts: n/a
Default

On 10 Apr 2004 12:03:21 -0700, (esara) wrote:

Accroding to this diagaram
http://www.viatech.com/en/k7-series/kt600.jsp

If for example My computer has DDR 400 and the CPU FSB is 400. In this
case I am wasting my memory speed. I mean the memory can send/receive
data to/from the chipset at speed 800Mhz(400*2) but the CPU which
suppose to handle these data can only work at FSB =400Mhz (I know the
internal speed is more than 400Mhz) but what I want to say is that
while the memory can supply data to the CPU at speed 800 the FSB of
the CPU can only take data at speed 400Mhz (so the FSB will slow down
the traffic). In this case this is not good design, the better is to
have CPU with FSB 800 or Why I bother to install memory DDR400, while
DDR200 will give the same performance provided that the CPU is FSB400.

Am I right?? any help would be very much apprciate it. Thanks.


Theoretically, you could have the same maximum bandwidth on that 2 X
DDR200. And you will have a better advantage of dual channel on slower
ram, of course. But latency will be slower with slower ram. And
there's the question of alignment of the channels. And actually, it's
latency that gives a general effect on your cpu application
performance. Bandwidth only affects performance during those instants
the application bottlenecks in memory bandwidth.
When, and if, that will be, is up to the application and the cpu. A
faster cpu would bottleneck more often, and needs faster buses.

ancra

  #7  
Old April 13th 04, 10:19 AM
Wes Newell
external usenet poster
 
Posts: n/a
Default

On Tue, 13 Apr 2004 00:35:58 +0200, somebody wrote:

On Sat, 10 Apr 2004 13:01:39 -0700, "Jim" wrote:

Ok, one problem here is that the marketing hype leads the advertising people
to not quite state things correctly. Let's forget the details of that ad,
and talk as accurately as possible.

The CPU FSB is more correctly stated as 200MHz (actual), but since it
employs DDR technology, it's *effective* rate is 400MHz. The "400MHz"
number you see in the ad is actually misleading, the CPU FSB does NOT
physically *run* at 400MHz, it runs at 200MHz (that's what would show up on
a scope!), *but*, because the DDR technology it employs allows data to
travel BOTH on the up and down side of each cycle (per MHz), it's *behaving*
as if it data was traveling on ONE side of the cycle, but at 400MHz! Get
it? It's a marketing gimic, the CPU isn't actually *faster*, it's more
*efficient* (2x in fact) at the same speed of a 200MHz processor (1x) that
does NOT employ DDR technology.


The clock is just that. It's just a clock. It might run at 'just' 200
MHz, but it's _NOT_ the speed of the bus! The speed of the bus is
400MHz, and there's no marketing gimmick about that. That is indeed
the _speed_!


Bus speed is measured by the clock cycles. The speed is 200MHz, not double
that or 4 times that. That is the data rate. Data rates are measured in
Bps or bps, not MHz. And anyone that thinks any different is just a stupid
stubborn idiot. Granted, that the bandwidth no longer depends on bus
speed, but data rate. The cpu clock speed does however depend on the bus
speed. Every cpu has a multiplier that sets the cpu clock speed in
accordance with the FSB speed. And now just a few problems that can arise
from calling the data rate the the FSB speed.

Tell an engineer to build you a MB with a 400MHz bus do you think they'll
know that you really only mean 100, or 200 in AMD's case.

Use that 400MHz to calculate your cpu speed and see how far off you'll be.

Now what purpose does the clock have? It's like this: When data is
transmitted on the FSB, data is represented on the 72 pins (64 + ecc)
out by voltage levels. But as the data have to change to a new value,
how do we know _when_ to read the data? When are all the leads ready,
and the data correct? That's where the clock comes in. The clock syncs
the transfer. And that is the _ONLY_ purpose of any clock. The clock
tells _when_ the data is supposed to be ready, and ok to read. Now, you
can sync on the clocks rising flank, or you can sync on the falling
flank. ... - Or, _both_!


A good example of DDR. If the clock speed doesn't mean anything, why does
the term DDR exist?:-)

AMD's FSB is the DEC Alpha EV6 protocol bus. And this happens to sync on
both rising and falling flanks.

So for a bus speed of 400MHz, - a 200MHz clock is _required_! The data
on the pins change 400 million times per second. That's the _SPEED_ of
bus! And that's only thing that matters, and marketing is entirely 100%
correct in stating that FSB speed is 400MHz. That is not marketing hype
or a gimmick.

But you are forgeting the major point. Bus speeds are mearsured by clock
speeds. Data speeds are measured by throughput in bps/Bps. Saying the bus
speed is 400MHz simply because the data rate is DDR is like saying my car
is going 120MPH when I have 2 people in it actually doing 60MPH.

I suppose I have to blame Intel marketing for everybody to be so damn
hung up on clockrates. - Hey, guys, - it's just a clock!

And without a clock, absolutely nothing in the system would work. I blame
both Intel and AMD for trying to change standard engineering practices by
multiplying apples (clock in MHz) times oranges (data in bps) and then
calling the final number apples. Pure BS.

As for DDR ram, I don't know how it works, but I assume the actual
transfer is something similar to the EV6 bus. But there are more complex
things involved with memory access. DDR speed only affects the bandwidth
that memoryblocks can be transferred with. An actual access is slower. A
long chain of things need to respond,

DDR ram is designated by bandwidth. PC2100 ram has a clock rate of 133MHz.
PC3200 has a clock rate of 200MHz, not 400 as you'll see all over the
f*cking place now.

And as my last comment on this. When I pinned AMD to the wall, they
admitted that their bogus FSB speeds were just that, by saying that it's
the effective clock rate when compared to a non DDR bus.

--
Abit KT7-Raid (KT133) Tbred B core CPU @2400MHz (24x100FSB)
http://mysite.verizon.net/res0exft/cpu.htm
  #8  
Old April 13th 04, 03:18 PM
Jim
external usenet poster
 
Posts: n/a
Default

You say TomAto, I say TOmatO.

Hey, I didn't create the nonemclature, and that's the problem. We don't
have any substitute for MHz, not like they now use w/ memory (PC2100,
PC2700, etc.). At least this is more honest in that it uses a common
denominator, bandwidth. But even here, clock matters, because THAT'S THE
WAY MOTHERBOARDS ARE CONFIGURED!

Tell ya what, explain to all of us why a P4 2.6C 800MHz CPU is synchronous
to PC3200 *without* mentioning clock speed. Now explain to us all the the
purposes and effects of CPU/DRAM ratios *without* mentioning clock speed.
Fact is, you can't! That's the lingua franca of configuration, like it or
not. If you don't understand and emphasize clock speed, NONE OF IT MAKES
SENSE! That's why we're all "hung up" on it!

You can't win in these forums. If you refer to 800MHz, you're pummeled by
posters as an idiot for not recognizing this is NOT the clock speed, it's
the clock that matters. If you refer to 200MHz, you're pummeled by the
other posters as an idiot for being "hung up" on clock speed, that's its the
throughput that matters. Christ's sake, you just can't win, you're damned
if you do, damned if you don't. That's why I usually ignore this dribble.
Just tired of being caught in the middle. I'm not here to help people pass
their MIT finals, I'm here to assist in understanding how to know what to
buy, and how to configure it. To this extent, my response was exactly
correct, indeed if I may be so bold, PERFECT! I wouldn't retract or restate
a damn thing.

As Wes put it so well, if YOU think that two people traveling in a car at 60
MPH means they're really traveling 120 MPH, and that's not misleading, well,
hey, you're entitled to you opinion. And I'm not about to lead the charge
on new nonemclature, it is what it is. We have no other common denominator
but MHz to compare processors, and more importantly, explain configuration.
When you've managed to sucessfully evangelize your new nomenclature, get
back to me.

Jim


wrote in message
...
On Sat, 10 Apr 2004 13:01:39 -0700, "Jim" wrote:

Ok, one problem here is that the marketing hype leads the advertising

people
to not quite state things correctly. Let's forget the details of that

ad,
and talk as accurately as possible.

The CPU FSB is more correctly stated as 200MHz (actual), but since it
employs DDR technology, it's *effective* rate is 400MHz. The "400MHz"
number you see in the ad is actually misleading, the CPU FSB does NOT
physically *run* at 400MHz, it runs at 200MHz (that's what would show up

on
a scope!), *but*, because the DDR technology it employs allows data to
travel BOTH on the up and down side of each cycle (per MHz), it's

*behaving*
as if it data was traveling on ONE side of the cycle, but at 400MHz! Get
it? It's a marketing gimic, the CPU isn't actually *faster*, it's more
*efficient* (2x in fact) at the same speed of a 200MHz processor (1x)

that
does NOT employ DDR technology.


The clock is just that. It's just a clock. It might run at 'just' 200
MHz, but it's _NOT_ the speed of the bus! The speed of the bus is
400MHz, and there's no marketing gimmick about that. That is indeed
the _speed_!
Now what purpose does the clock have?
It's like this: When data is transmitted on the FSB, data is
represented on the 72 pins (64 + ecc) out by voltage levels.
But as the data have to change to a new value, how do we know _when_
to read the data? When are all the leads ready, and the data correct?
That's where the clock comes in. The clock syncs the transfer. And
that is the _ONLY_ purpose of any clock. The clock tells _when_ the
data is supposed to be ready, and ok to read.
Now, you can sync on the clocks rising flank, or you can sync on the
falling flank. ... - Or, _both_!
AMD's FSB is the DEC Alpha EV6 protocol bus. And this happens to sync
on both rising and falling flanks.

So for a bus speed of 400MHz, - a 200MHz clock is _required_!
The data on the pins change 400 million times per second. That's the
_SPEED_ of bus! And that's only thing that matters, and marketing is
entirely 100% correct in stating that FSB speed is 400MHz.
That is not marketing hype or a gimmick.

I suppose I have to blame Intel marketing for everybody to be so damn
hung up on clockrates. - Hey, guys, - it's just a clock!

As for DDR ram, I don't know how it works, but I assume the actual
transfer is something similar to the EV6 bus. But there are more
complex things involved with memory access.
DDR speed only affects the bandwidth that memoryblocks can be
transferred with. An actual access is slower. A long chain of things
need to respond,

AMD did the same thing w/ their processors. Is an AMD
Athlon 2600+ actually running at 2.6GHz? No way, that's the hype, it's
actually 2.08GHz "on scope", AMD is merely claiming 2600 (2.6GHz) is the
*effective* performance compared to a 2.6GHz Intel CPU. Fact or fiction?
You decide.


It's not quite the same thing. But it's the same in the sense that the
clockrate is not the "speed" of any cpu. A clock is still just a
clock. The clock doesn't do the actual work.
Since Intel went for misleading the market with the P4, -

(The P4 is unique, since it's _less_ efficient per clockcycle than
preceding cpus, Pentium, PII, PIII. All other new generation cpus,
have always been more efficient per clockcycle than previous. That's a
natural outcome of an increased number of transistors. In the case of
the P4 and Prescott, their increased transistor counts are dedicated
to make it possible to run a higher clock. Not to do more work. This
is not good engineering. The P4 is engineered, intentionally, to run
at as high clock as possible. Certainly sacrificing performance! And
_THIS_ is indeed entirely marketing hype. A "gimic".)

, - AMD went for a 'rating' naming system instead. ...And so will
Intel shortly...

ancra



  #9  
Old April 13th 04, 10:11 PM
external usenet poster
 
Posts: n/a
Default

On Tue, 13 Apr 2004 07:18:42 -0700, "Jim" wrote:

You say TomAto, I say TOmatO.

Hey, I didn't create the nonemclature, and that's the problem. We don't
have any substitute for MHz, not like they now use w/ memory (PC2100,
PC2700, etc.). At least this is more honest in that it uses a common
denominator, bandwidth. But even here, clock matters, because THAT'S THE
WAY MOTHERBOARDS ARE CONFIGURED!

Tell ya what, explain to all of us why a P4 2.6C 800MHz CPU is synchronous
to PC3200 *without* mentioning clock speed. Now explain to us all the the
purposes and effects of CPU/DRAM ratios *without* mentioning clock speed.
Fact is, you can't! That's the lingua franca of configuration, like it or
not. If you don't understand and emphasize clock speed, NONE OF IT MAKES
SENSE! That's why we're all "hung up" on it!

You can't win in these forums. If you refer to 800MHz, you're pummeled by
posters as an idiot for not recognizing this is NOT the clock speed, it's
the clock that matters. If you refer to 200MHz, you're pummeled by the
other posters as an idiot for being "hung up" on clock speed, that's its the
throughput that matters. Christ's sake, you just can't win, you're damned
if you do, damned if you don't. That's why I usually ignore this dribble.
Just tired of being caught in the middle. I'm not here to help people pass
their MIT finals, I'm here to assist in understanding how to know what to
buy, and how to configure it. To this extent, my response was exactly
correct, indeed if I may be so bold, PERFECT! I wouldn't retract or restate
a damn thing.

As Wes put it so well, if YOU think that two people traveling in a car at 60
MPH means they're really traveling 120 MPH, and that's not misleading, well,
hey, you're entitled to you opinion. And I'm not about to lead the charge
on new nonemclature, it is what it is. We have no other common denominator
but MHz to compare processors, and more importantly, explain configuration.
When you've managed to sucessfully evangelize your new nomenclature, get
back to me.


Yes Jim. Your response was exactly correct. Rereading it I realized
that too.

I was kind of reacting to thinking that 'effective FSB speed' was
portrayed as some kind of falsification.
If some kind of apology is required, you have it.
Still, this issue needs discussion, so I'm not sorry for my post.

But I'm not in the habit of comparing processors by MHz, and I'm sure,
neither are you :-D (got you there ;-))

As for cars and MPH, check out my response to Wes.

ancra

  #10  
Old April 13th 04, 10:24 PM
external usenet poster
 
Posts: n/a
Default

On Tue, 13 Apr 2004 09:19:18 GMT, Wes Newell
wrote:

On Tue, 13 Apr 2004 00:35:58 +0200, somebody wrote:

On Sat, 10 Apr 2004 13:01:39 -0700, "Jim" wrote:

Ok, one problem here is that the marketing hype leads the advertising people
to not quite state things correctly. Let's forget the details of that ad,
and talk as accurately as possible.

The CPU FSB is more correctly stated as 200MHz (actual), but since it
employs DDR technology, it's *effective* rate is 400MHz. The "400MHz"
number you see in the ad is actually misleading, the CPU FSB does NOT
physically *run* at 400MHz, it runs at 200MHz (that's what would show up on
a scope!), *but*, because the DDR technology it employs allows data to
travel BOTH on the up and down side of each cycle (per MHz), it's *behaving*
as if it data was traveling on ONE side of the cycle, but at 400MHz! Get
it? It's a marketing gimic, the CPU isn't actually *faster*, it's more
*efficient* (2x in fact) at the same speed of a 200MHz processor (1x) that
does NOT employ DDR technology.


The clock is just that. It's just a clock. It might run at 'just' 200
MHz, but it's _NOT_ the speed of the bus! The speed of the bus is
400MHz, and there's no marketing gimmick about that. That is indeed
the _speed_!


Bus speed is measured by the clock cycles.


What is that "Bus speed"? And why is it "measured by the clock
cycles"?
Who does this? - Certainly not AMD or Intel.
That's just the thing here. If that "bus speed" is measured in clock
cycles, then it's because that "bus speed" is the bus clock speed.
We're running in circles.

The speed is 200MHz, not double
that or 4 times that. That is the data rate. Data rates are measured in
Bps or bps, not MHz.


Frequency of data transfers should be measured in MHz (Hz), just like
any frequency of any event. Clocks do not own the unit MHz.
But sure, data rate is fine to measure in Bps.

And anyone that thinks any different is just a stupid
stubborn idiot.


- Tsk, tsk. - I'll reserve my judgement on the making of such a
'statement'. :-D

Granted, that the bandwidth no longer depends on bus
speed, but data rate. The cpu clock speed does however depend on the bus
speed. Every cpu has a multiplier that sets the cpu clock speed in
accordance with the FSB speed. And now just a few problems that can arise
from calling the data rate the the FSB speed.


You can forget about stupid, and idiot too, - well, at least in the IQ
sense.
But: - Stubborn? - Oh most certainly. I'm not married!
So here goes:

The bandwidth does depend on a 'speed'. How could it possibly not?
And the 'effective FSB speed' is not the same thing as data rate.
Data rate is width of bus times 'effective bus speed'.

(And I could disagree, if I wanted to, what we see are some few
problems arising from calling 'FSB clock speed' 'FSB speed' ;-).)

'Bus speed' is an unfortunate term, since AMD and Intel use it as
short for 'effective bus speed', in the case of FSB, while it's also
established in use as short for 'bus clock speed'. I only used that
term (bus speed) once in my post, and I propose that we henceforth are
explicit about this and use 'effective FSB speed' respectively 'FSB
clock speed', or our argument will indeed be stupid.
So ok, I agree, maybe marketing haven't made us any immediate favor
here.

I think I primarily used the term 'speed'. Also as in 'speed of the
bus', which you interpret as 'clockspeed' at every turn. Please don't
do that. I meant speed as speed. As does AMD and Intel.
"Speed" is speed. 'Clock speed' is just a term for the frequency of
the clock pulses. It doesn't own 'speed'.

Tell an engineer to build you a MB with a 400MHz bus do you think they'll
know that you really only mean 100, or 200 in AMD's case.


- But isn't that pretty much exactly what DEC did? "Give us a 200MHz
bus" and they came up with the Alpha bus at a 100MHz clock?
I would think anyone actually asking for a 'speed'(sic) is concerned
with speed in terms of performance.
(And I think any engineers building a mobo would have to work from
very detailed specs, not just a loose MHz figure.)

Use that 400MHz to calculate your cpu speed and see how far off you'll be.


Why should I? If I know enough to calculate cpu 'clock speed', I'd
know I need a multiplier and an external _CLOCK_. Why should I grasp
blindly for any and all MHz figure floating around?

The DDR rates and "effective bus speed" again in MHz, are terms and
concepts established in language and specs. It might not be to your
liking, but they are technically motivated, and insisting on your ways
is not very constructive. The concepts needs explaining, not
dismissing. To more often strictly employ 'data rate' and 'Bps' is
excellent. I think you're right about that (still there's a
possibility for another mixup with 'data transfer rate' here). But
language develops as language will. There's not terribly much we can
do about it, but explain and try to avoid misunderstandings. Not using
'speed' when you really mean 'clock speed' is also helpful.

Language also evolves when there's specific communication needs,
within specific groups. In this particular case the need is to
communicate the relevant properties to users. The simplistic concept
of always understanding 'speed' exclusively as clockrate might work
for a digital engineer with his arms full of detailed manuals. But he
doesn't need that. He can make do with the more explicit 'clock
speed'. It's also not proper for the market. The market assumes
'speed' denotes an aspect of performance, since that is usually
fundamental to the concept of speed. So there is a need for concepts
like "effective FSB speed" and DDR rates.

Since that "400MHz" FSB is an established concept, there is the need
to explain it's not the clock. Likewise, there is the need to explain
why we could possibly have use for PC2700 on a 166MHz FSB clock (for
all those assuming 'clock speed' is the speed(sic) of the bus).

SNIP

But you are forgeting the major point. Bus speeds are mearsured by clock
speeds. Data speeds are measured by throughput in bps/Bps. Saying the bus
speed is 400MHz simply because the data rate is DDR is like saying my car
is going 120MPH when I have 2 people in it actually doing 60MPH.


Bus speed is the movement rate over ground, of a large vehicle,
transporting people, traveling on the road, and is measured in mph or
km/h... ;-)
'Bus clock speed' or 'effective bus speed' on the other hand...

- And I don't agree, - it's more like saying your car is doing 60mph
at 6000rpm, in 2'nd gear, when it is in fact indeed traveling at
60mph, even though it would only do 30mph at the same 6000rpm in 1'st
gear.
(And if those figures are totally wild off, I'm sorry, I don't know
much about cars, you know what I mean any way.)

Suppose we had only one gear, and an established tradition of
expressing speed of a vehicle, in terms of engine rpm. And also
suppose we had a tradition of interpreting 'speed' as short for
'engine speed', how appropriate will all that be when someone invents
the gearbox?
And how silly is it to be frantically wailing that it's more "correct"
to say that the 'speed' of the car is 6000rpm?
(I'm not saying you're actually doing that, but you could be ;-))

(The number of persons traveling in the car looks more like width of
the bus, or dual channels, to me.)

Frequency of pulses on a clock is measured in MHz.
Frequency of data transfers occurring on a bus is also measured in
MHz.
I'm perfectly aware of which one of these is the 'clock speed' or 'bus
clock speed'.
The transport capacity of the bus is of course the throughput, in Bps,
but wouldn't that also be the width of the bus times it's *speed* or,
more exactly, it's 'effective speed'?
- So, what is the speed(sic) of the bus?

Is the transport capacity, of a vehicle, the number of seats times
engine rpm?

And without a clock, absolutely nothing in the system would work. I blame
both Intel and AMD for trying to change standard engineering practices by
multiplying apples (clock in MHz) times oranges (data in bps) and then
calling the final number apples. Pure BS.


Yes, I can possibly understand some of those sentiments. But
technology and the world changes. And so there are some people that
feel the need for new terms, since perceptions of properties, commonly
associated with the old terms, are no longer valid. No one is in a
better position than AMD and Intel for that. That's how it goes. Chew
well.

However, In this case I'd also argue it's not a case of "apples times
oranges". MHz is not an exclusive unit for measuring clockrates,
neither is 'speed' the exclusive property of clock frequencies. MHz is
a unit for measuring frequency of an event. In the case of 'effective
bus speed', I've always assumed it's the frequency of data transfers.
It makes sense to me.

My point, which I possibly was somewhat overanxious to get across, (by
responding to the previous posters perfectly correct post) was that
there is nothing 'write up' or 'fake' about 'effective fsb speed', in
terms of relevant properties. On the contrary, it's the 'bus clock
speed' that carries the wrong implications. Particularly if you're in
a habit of calling it 'bus speed'.

SNIP

DDR ram is designated by bandwidth. PC2100 ram has a clock rate of 133MHz.
PC3200 has a clock rate of 200MHz, not 400 as you'll see all over the
f*cking place now.


Yes, DDR ram modules are designated by maximum bandwidth, it's max
transfer speed (sic) is designated by its DDR400 rating, and the
clockspeed for that is half (as implied by "double..."). I certainly
know all that! That is not what I was considering when I said "I don't
know how it works". But never mind.
It should also be obvious that the '400', that you're so annoyed with,
is technically motivated, since it denotes the transfer rate,
irrespectively of the width of the module..

And as my last comment on this. When I pinned AMD to the wall, they
admitted that their bogus FSB speeds were just that, by saying that it's
the effective clock rate when compared to a non DDR bus.


I'm pretty sure though, that they didn't admit to any "bogus"? But
rather tried to explain what it meant, in terms of relevant
properties?

ancra

 




Thread Tools
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

vB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Forum Jump

Similar Threads
Thread Thread Starter Forum Replies Last Post
On the brink of madness... I.C. Koets General 18 January 31st 05 11:49 PM
Athlon 64 queries Spiro AMD x86-64 Processors 11 September 19th 04 01:30 AM
OC on Asus P4P 800 redrider Overclocking 8 March 11th 04 09:02 AM
FSB settings increase memory speed? GlueGum Overclocking AMD Processors 2 February 4th 04 02:35 AM
Memory timings for best speed Lee Overclocking 0 July 2nd 03 11:14 PM


All times are GMT +1. The time now is 10:49 PM.


Powered by vBulletin® Version 3.6.4
Copyright ©2000 - 2024, Jelsoft Enterprises Ltd.
Copyright ©2004-2024 HardwareBanter.
The comments are property of their posters.