The True FSB???

[Guest]

Archived from groups: alt.comp.hardware.overclocking,alt.comp.hardware.overclocking.amd (More info?)

To an Intel P4.

100 MHz
133 MHz (133.25)
200 MHz
266 MHz (266.5)
and soon 512 MHz
Even Intel's Processor identification Utility lists like this...


Intel(R) Processor Identification Utility
Version: 1.5.20050202
Time Stamp: 2005/04/03 09:06:08
Number of processors in system: 2
Current processor: #1
Processor Name: Intel(R) Xeon(TM) CPU 2.66GHz
Type: 0
Family: F
Model: 2
Stepping: 5
Revision: 11
L1 Trace Cache: 12 Kµops
L1 Data Cache: 8 KB
L2 Cache: 512 KB
L3 Cache: None
Packaging: OOI
EIST: No
MMX(TM): Yes
SIMD: Yes
SIMD2: Yes
SIMD3: No
Enhanced Halt State: No
Execute Disable Bit: No
Hyper-Threading Technology: Yes
Intel(R) Extended Memory 64 Technology: No
Expected Processor Frequency: 2.66 GHz
Reported Processor Frequency: 3.20 GHz
Expected System Bus Frequency: 533 MHz
Reported System Bus Frequency: 640 MHz

I don't see the words...
Front Side Bus @ all.

RavingRaichu. ;-)

 

[Guest]

Archived from groups: alt.comp.hardware.overclocking,alt.comp.hardware.overclocking.amd (More info?)

On Sun, 03 Apr 2005 05:08:46 -0400, RaichuRaichu wrote:

> To an Intel P4.
>
> 100 MHz
> 133 MHz (133.25)
> 200 MHz
> 266 MHz (266.5)
> and soon 512 MHz

I wouldn't expect to see a 512MHz FBS anytime soon. The 266 Just came out.
Besides, it wouldn't be 512. either 500 or 533. But before they get there
they would use something in between. But don't count on it anytime soon,
if ever within the next 5 years.

> Even Intel's Processor identification Utility lists like this...
>
> Expected Processor Frequency: 2.66 GHz
> Reported Processor Frequency: 3.20 GHz
> Expected System Bus Frequency: 533 MHz
> Reported System Bus Frequency: 640 MHz
>
> I don't see the words...
> Front Side Bus @ all.
>
They are not being specific and they are calling it the system bus. But
that doesn't matter either, because the true FSB speed for this
overclocked cpu is 160MHz, quad pumped data rate (for 640Mbps x bus
width), or QDR640 as I coined it.

I'm not sure what your question was but mabye that answered it.

--
Abit KT7-Raid (KT133) Tbred B core CPU @2400MHz (24x100FSB)
My server http://wesnewell.no-ip.com/cpu.php
Verizon server http://mysite.verizon.net/res0exft/cpu.htm

 

[Guest]

Archived from groups: alt.comp.hardware.overclocking,alt.comp.hardware.overclocking.amd (More info?)

Wes Newell wrote:
> On Sun, 03 Apr 2005 05:08:46 -0400, RaichuRaichu wrote:
>
>
>>To an Intel P4.
>>
>>100 MHz
>>133 MHz (133.25)
>>200 MHz
>>266 MHz (266.5)
>>and soon 512 MHz
>
>
> I wouldn't expect to see a 512MHz FBS anytime soon. The 266 Just came out.
> Besides, it wouldn't be 512. either 500 or 533. But before they get there
> they would use something in between. But don't count on it anytime soon,
> if ever within the next 5 years.
>
>
>>Even Intel's Processor identification Utility lists like this...
>>
>>Expected Processor Frequency: 2.66 GHz
>>Reported Processor Frequency: 3.20 GHz
>>Expected System Bus Frequency: 533 MHz
>>Reported System Bus Frequency: 640 MHz
>>
>>I don't see the words...
>>Front Side Bus @ all.
>>
>
> They are not being specific and they are calling it the system bus. But
> that doesn't matter either, because the true FSB speed for this
> overclocked cpu is 160MHz,

You mean the FSB "clock" is 160 MHz.

> quad pumped data rate (for 640Mbps x bus
> width), or QDR640 as I coined it.
>
> I'm not sure what your question was but mabye that answered it.
>

 

[Guest]

Archived from groups: alt.comp.hardware.overclocking,alt.comp.hardware.overclocking.amd (More info?)

RaichuRaichu wrote:

> To an Intel P4.
>
> 100 MHz
> 133 MHz (133.25)
> 200 MHz
> 266 MHz (266.5)

Those are bus clocks. Although 133 is really 133.33... not 133.25, and 266
is 266.66... not 266.5.

> and soon 512 MHz
> Even Intel's Processor identification Utility lists like this...
>
>
> Intel(R) Processor Identification Utility
> Version: 1.5.20050202
> Time Stamp: 2005/04/03 09:06:08
> Number of processors in system: 2
> Current processor: #1
> Processor Name: Intel(R) Xeon(TM) CPU 2.66GHz
> Type: 0
> Family: F
> Model: 2
> Stepping: 5
> Revision: 11
> L1 Trace Cache: 12 Kµops
> L1 Data Cache: 8 KB
> L2 Cache: 512 KB
> L3 Cache: None
> Packaging: OOI
> EIST: No
> MMX(TM): Yes
> SIMD: Yes
> SIMD2: Yes
> SIMD3: No
> Enhanced Halt State: No
> Execute Disable Bit: No
> Hyper-Threading Technology: Yes
> Intel(R) Extended Memory 64 Technology: No
> Expected Processor Frequency: 2.66 GHz
> Reported Processor Frequency: 3.20 GHz
> Expected System Bus Frequency: 533 MHz
> Reported System Bus Frequency: 640 MHz
>
> I don't see the words...
> Front Side Bus @ all.

So?

>
> RavingRaichu. ;-)
>
>

 

[Guest]

Archived from groups: alt.comp.hardware.overclocking,alt.comp.hardware.overclocking.amd (More info?)

"RaichuRaichu" Becomes very tiresome, and gets this suggestion:

Just to give you something to do; try various URLs such as
http://www.intel.com/design/xeon/datashts/252135.htm
(which will direct you to a document titled:


Intel(R) Xeon(TM) Processor with 533 MHz Front Side Bus at 2 GHz to 3.20 GHz
Datasheet.

This time, please read the document at the URL, which I've posted twice
before when you offered your stubborn insistence on knowing the one true
FSB.

And remember, just because you can find something in the Intel website
doesn't mean you should take it at face value. Intelligence on your part is
also necessary.

Phil Weldon

"RaichuRaichu" <ravingraichu@comcast.net> wrote in message
news:et2dnWfifcE_L9LfRVn-og@comcast.com...
> To an Intel P4.
>
> 100 MHz
> 133 MHz (133.25)
> 200 MHz
> 266 MHz (266.5)
> and soon 512 MHz
> Even Intel's Processor identification Utility lists like this...
>
>
> Intel(R) Processor Identification Utility
> Version: 1.5.20050202
> Time Stamp: 2005/04/03 09:06:08
> Number of processors in system: 2
> Current processor: #1
> Processor Name: Intel(R) Xeon(TM) CPU 2.66GHz
> Type: 0
> Family: F
> Model: 2
> Stepping: 5
> Revision: 11
> L1 Trace Cache: 12 Kµops
> L1 Data Cache: 8 KB
> L2 Cache: 512 KB
> L3 Cache: None
> Packaging: OOI
> EIST: No
> MMX(TM): Yes
> SIMD: Yes
> SIMD2: Yes
> SIMD3: No
> Enhanced Halt State: No
> Execute Disable Bit: No
> Hyper-Threading Technology: Yes
> Intel(R) Extended Memory 64 Technology: No
> Expected Processor Frequency: 2.66 GHz
> Reported Processor Frequency: 3.20 GHz
> Expected System Bus Frequency: 533 MHz
> Reported System Bus Frequency: 640 MHz
>
> I don't see the words...
> Front Side Bus @ all.
>
> RavingRaichu. ;-)
>
>

 

[Guest]

Archived from groups: alt.comp.hardware.overclocking,alt.comp.hardware.overclocking.amd (More info?)

"Phil Weldon" wrote in message...
> "RaichuRaichu" Becomes very tiresome...
>
> And remember, just because you can find something in the Intel website
> doesn't mean you should take it at face value. Intelligence on your part
> is also necessary.

....And therein lies the crux of this thread (and several others)...
--


Richard Hopkins
Cardiff, Wales, United Kingdom
(replace nospam with pipex in reply address)

The UK's leading technology reseller www.dabs.com

 

[Guest]

Archived from groups: alt.comp.hardware.overclocking,alt.comp.hardware.overclocking.amd (More info?)

"RaichuRaichu" wrote some more stuff about clocks and buses,
> I don't see the words...
> Front Side Bus @ all.

so just to broaden his prospective, this quote:

"This month, we scrutinize three new motherboards using Intel's 925XE
chipset, which supports the 1066MHz front-side bus."
"MaximumPC"
April 2005
page 84

Phil Weldon

"RaichuRaichu" <ravingraichu@comcast.net> wrote in message
news:et2dnWfifcE_L9LfRVn-og@comcast.com...
> To an Intel P4.
>
> 100 MHz
> 133 MHz (133.25)
> 200 MHz
> 266 MHz (266.5)
> and soon 512 MHz
> Even Intel's Processor identification Utility lists like this...
>
>
> Intel(R) Processor Identification Utility
> Version: 1.5.20050202
> Time Stamp: 2005/04/03 09:06:08
> Number of processors in system: 2
> Current processor: #1
> Processor Name: Intel(R) Xeon(TM) CPU 2.66GHz
> Type: 0
> Family: F
> Model: 2
> Stepping: 5
> Revision: 11
> L1 Trace Cache: 12 Kµops
> L1 Data Cache: 8 KB
> L2 Cache: 512 KB
> L3 Cache: None
> Packaging: OOI
> EIST: No
> MMX(TM): Yes
> SIMD: Yes
> SIMD2: Yes
> SIMD3: No
> Enhanced Halt State: No
> Execute Disable Bit: No
> Hyper-Threading Technology: Yes
> Intel(R) Extended Memory 64 Technology: No
> Expected Processor Frequency: 2.66 GHz
> Reported Processor Frequency: 3.20 GHz
> Expected System Bus Frequency: 533 MHz
> Reported System Bus Frequency: 640 MHz
>
> I don't see the words...
> Front Side Bus @ all.
>
> RavingRaichu. ;-)
>
>

 

[Guest]

Archived from groups: alt.comp.hardware.overclocking,alt.comp.hardware.overclocking.amd (More info?)

"David Maynard" <nospam@private.net> wrote in message
news:1150mh6cjc3t58c@corp.supernews.com...
RaichuRaichu wrote:

> To an Intel P4.
>
> 100 MHz
> 133 MHz (133.25)
> 200 MHz
> 266 MHz (266.5)

Those are bus clocks. Although 133 is really 133.33... not 133.25, and 266
is 266.66... not 266.5.

Yeah, that's really very true.
This thing is rated @ like 2,666.6 MHz I don't remember all the numbers.

Of course, it's easy to figure out where I got those numbers.

And the main Point.
The True Front Side Bus has got so lost even some major companys
don't even know which is (truly) which.
> and soon 512 MHz
> Even Intel's Processor identification Utility lists like this...
>
>
> Intel(R) Processor Identification Utility
> Version: 1.5.20050202
> Time Stamp: 2005/04/03 09:06:08
> Number of processors in system: 2
> Current processor: #1
> Processor Name: Intel(R) Xeon(TM) CPU 2.66GHz
> Type: 0
> Family: F
> Model: 2
> Stepping: 5
> Revision: 11
> L1 Trace Cache: 12 Kµops
> L1 Data Cache: 8 KB
> L2 Cache: 512 KB
> L3 Cache: None
> Packaging: OOI
> EIST: No
> MMX(TM): Yes
> SIMD: Yes
> SIMD2: Yes
> SIMD3: No
> Enhanced Halt State: No
> Execute Disable Bit: No
> Hyper-Threading Technology: Yes
> Intel(R) Extended Memory 64 Technology: No
> Expected Processor Frequency: 2.66 GHz
> Reported Processor Frequency: 3.20 GHz
> Expected System Bus Frequency: 533 MHz
> Reported System Bus Frequency: 640 MHz
>
> I don't see the words...
> Front Side Bus @ all.

So?

>
> RavingRaichu. ;-)
>
>

 

[Guest]

Archived from groups: alt.comp.hardware.overclocking,alt.comp.hardware.overclocking.amd (More info?)

"Wes Newell" <w.newell@TAKEOUTverizon.net> wrote in message
newsan.2005.04.03.11.32.52.512239@TAKEOUTverizon.net...
> On Sun, 03 Apr 2005 05:08:46 -0400, RaichuRaichu wrote:
>
> > To an Intel P4.
> >
> > 100 MHz
> > 133 MHz (133.25)
> > 200 MHz
> > 266 MHz (266.5)
> > and soon 512 MHz
>
> I wouldn't expect to see a 512MHz FBS anytime soon. The 266 Just came out.
> Besides, it wouldn't be 512. either 500 or 533. But before they get there
> they would use something in between. But don't count on it anytime soon,
> if ever within the next 5 years.
A Computer shop owner I was talking to when I said about
AMD's FX55 @ 1,000 FSB said he had an Intel 775 (LGA)@
2048 MHz FSB. 2.8GHz Or split to 4 for the 512 (True) FSB
Then told me that AMD only had an 800 MHz FSB (400) Really.
When I tried to point out the true FSB, he jumped on my case with
MCSE NOT A+
Though I don't know all the new stuff as well as I could, I do know most
of the other older stuff well enough to pass an A+ cert.
If I know things as well as I think I know.
MCSE is just knowing OS very well, while A+ Cert
is knowing hardware.
Of course I just pointed out now AMD's nice chip.
Even @ 3.8 GHz, the P4 cant keep up with a system bus of 2,000 MHz (2GHz)
What more I want to say about this is better in a new thread. (LOL)

RavingRaichu. ;-) (Denny.)
>
> > Even Intel's Processor identification Utility lists like this...
> >
> > Expected Processor Frequency: 2.66 GHz
> > Reported Processor Frequency: 3.20 GHz
> > Expected System Bus Frequency: 533 MHz
> > Reported System Bus Frequency: 640 MHz
> >
> > I don't see the words...
> > Front Side Bus @ all.
> >
> They are not being specific and they are calling it the system bus. But
> that doesn't matter either, because the true FSB speed for this
> overclocked cpu is 160MHz, quad pumped data rate (for 640Mbps x bus
> width), or QDR640 as I coined it.
>
> I'm not sure what your question was but mabye that answered it.
>
> --
> Abit KT7-Raid (KT133) Tbred B core CPU @2400MHz (24x100FSB)
> My server http://wesnewell.no-ip.com/cpu.php
> Verizon server http://mysite.verizon.net/res0exft/cpu.htm
>

 

[Guest]

Archived from groups: alt.comp.hardware.overclocking,alt.comp.hardware.overclocking.amd (More info?)

RaichuRaichu wrote:

> "David Maynard" <nospam@private.net> wrote in message
> news:1150mh6cjc3t58c@corp.supernews.com...
> RaichuRaichu wrote:
>
>
>>To an Intel P4.
>>
>>100 MHz
>>133 MHz (133.25)
>>200 MHz
>>266 MHz (266.5)
>
>
> Those are bus clocks. Although 133 is really 133.33... not 133.25, and 266
> is 266.66... not 266.5.
>
> Yeah, that's really very true.
> This thing is rated @ like 2,666.6 MHz I don't remember all the numbers.
>
> Of course, it's easy to figure out where I got those numbers.
>
> And the main Point.
> The True Front Side Bus has got so lost even some major companys
> don't even know which is (truly) which.

The Front Side bus is not 'lost' and never has been.

>
>>and soon 512 MHz
>>Even Intel's Processor identification Utility lists like this...
>>
>>
>>Intel(R) Processor Identification Utility
>>Version: 1.5.20050202
>>Time Stamp: 2005/04/03 09:06:08
>>Number of processors in system: 2
>>Current processor: #1
>>Processor Name: Intel(R) Xeon(TM) CPU 2.66GHz
>>Type: 0
>>Family: F
>>Model: 2
>>Stepping: 5
>>Revision: 11
>>L1 Trace Cache: 12 Kµops
>>L1 Data Cache: 8 KB
>>L2 Cache: 512 KB
>>L3 Cache: None
>>Packaging: OOI
>>EIST: No
>>MMX(TM): Yes
>>SIMD: Yes
>>SIMD2: Yes
>>SIMD3: No
>>Enhanced Halt State: No
>>Execute Disable Bit: No
>>Hyper-Threading Technology: Yes
>>Intel(R) Extended Memory 64 Technology: No
>>Expected Processor Frequency: 2.66 GHz
>>Reported Processor Frequency: 3.20 GHz
>>Expected System Bus Frequency: 533 MHz
>>Reported System Bus Frequency: 640 MHz
>>
>>I don't see the words...
>>Front Side Bus @ all.
>
>
> So?
>
>
>>RavingRaichu. ;-)
>>
>>
>
>
>
>

 

[Guest]

Archived from groups: alt.comp.hardware.overclocking,alt.comp.hardware.overclocking.amd (More info?)

"RaichuRaichu" <ravingraichu@comcast.net> wrote in message...
> The True Front Side Bus has got so lost even some major companys
> don't even know which is (truly) which.

Rofl. No Denny. The major companys <sic> definitely know which is which. So
do the rest of us. Seems the only one round here who's confused is you.

>> Even Intel's Processor identification Utility lists like this...
>> I don't see the words...
>> Front Side Bus @ all.
>
> So?

You don't see the term "front side bus" used in Intel's utility because
Intel would rather you use their new "System Bus Frequency" buzz-phrase
instead. Don't forget that "Front Side Bus" was itself originally an Intel
marketing phrase; they've just invented another one to supercede it.

>> RavingRaichu. ;-)

Well, at least one of these words is accurate. Keep taking the tablets
Denny.
--


Richard Hopkins
Cardiff, Wales, United Kingdom
(replace nospam with pipex in reply address)

The UK's leading technology reseller www.dabs.com

 

[Guest]

Archived from groups: alt.comp.hardware.overclocking,alt.comp.hardware.overclocking.amd (More info?)

On Mon, 04 Apr 2005 11:58:55 -0400, RaichuRaichu wrote:

> A Computer shop owner I was talking to when I said about
> AMD's FX55 @ 1,000 FSB said he had an Intel 775 (LGA)@
> 2048 MHz FSB. 2.8GHz Or split to 4 for the 512 (True) FSB

Just proves that computer shop owners can be ignorant too. The default FSB
speed of the FX 55 is the same as all the other K8 cpu's, 200MHz. But it's
been overclocked to more than 300Mhz by a few. 1000Mbps (not MHz) is the
dase data rate of the bus. So what he really did was to overclock the FSB
to 512MHz and then took the data rate of 2048Mbps and changed it to MHz.
Typical of someone that really doesn't know what they are doing.

> Then told me that AMD only had an 800 MHz FSB (400) Really.
> When I tried to point out the true FSB, he jumped on my case with
> MCSE NOT A+.

I had to show the first MCSE guy I hired how to wire up an ethernet
cable. I don't know what an A+ cert means and it really doesn't matter
much. Anyone can be ignorant, regardless of how many pieces of paper they
have that say they are trained. Trained does not equal smart. And I've
trained enough people to know this for a fact.

--
Abit KT7-Raid (KT133) Tbred B core CPU @2400MHz (24x100FSB)
My server http://wesnewell.no-ip.com/cpu.php
Verizon server http://mysite.verizon.net/res0exft/cpu.htm

 

[Guest]

Archived from groups: alt.comp.hardware.overclocking,alt.comp.hardware.overclocking.amd (More info?)

Wes Newell wrote:
> On Mon, 04 Apr 2005 11:58:55 -0400, RaichuRaichu wrote:
>
>> A Computer shop owner I was talking to when I said about
>> AMD's FX55 @ 1,000 FSB said he had an Intel 775 (LGA)@
>> 2048 MHz FSB. 2.8GHz Or split to 4 for the 512 (True) FSB
>
> Just proves that computer shop owners can be ignorant too. The
> default FSB speed of the FX 55 is the same as all the other K8 cpu's,
> 200MHz.

Denny's actually somewhat right on this point. The HT bus on an K8 does run
at 800MHz or 1000MHz, and uses DDR on top of that (at 16 bits wide each way
to get the 3.2 or 4 GBytes/sec bandwidth). The 200MHz is just used as a
common base clock for both the HT bus and the CPU speed.

[...]

--
Michael Brown
www.emboss.co.nz : OOS/RSI software and more
Add michael@ to emboss.co.nz ---+--- My inbox is always open

 

[Guest]

Archived from groups: alt.comp.hardware.overclocking,alt.comp.hardware.overclocking.amd (More info?)

On Tue, 05 Apr 2005 08:49:19 +1000, Michael Brown wrote:

> Wes Newell wrote:
>> On Mon, 04 Apr 2005 11:58:55 -0400, RaichuRaichu wrote:
>>
>>> A Computer shop owner I was talking to when I said about
>>> AMD's FX55 @ 1,000 FSB said he had an Intel 775 (LGA)@
>>> 2048 MHz FSB. 2.8GHz Or split to 4 for the 512 (True) FSB
>>
>> Just proves that computer shop owners can be ignorant too. The
>> default FSB speed of the FX 55 is the same as all the other K8 cpu's,
>> 200MHz.
>
> Denny's actually somewhat right on this point. The HT bus on an K8 does run
> at 800MHz or 1000MHz, and uses DDR on top of that (at 16 bits wide each way
> to get the 3.2 or 4 GBytes/sec bandwidth). The 200MHz is just used as a
> common base clock for both the HT bus and the CPU speed.
>
> [...]

Yep, I guess I kind of blew that part.

I've noticed that they've also raised to the allowable HT clock rate up to
1.4GHz since I last looked. I'm not sure why there's so much focus on
this though (for the FSB). With a seperate memory controller, the A64
doesn't need all of that bandwidth for other IO. For those interested;

14. At what clock speeds does HyperTransport technology operate?
HyperTransport technology devices are designed to operate at multiple
clock speeds from 200MHz up to 1.4 GHz, and utilize double data rate
technology transferring two bits of data per clock cycle, for an effective
transfer rate of up to 2.8 gigatransfer/sec in each direction. Since
transfers can occur in both directions simultaneously, an aggregate
transfer rate of 11.2 gigabytes per second in a 16 bit HyperTransport I/O
Link and an aggregate transfer rate of 22.4 gigabytes per second in a
32-bit HyperTransport I/O Link can be achieved. To allow for system design
optimization, the clocks of the receive and transmit links may beset at
different rates.

http://www.hypertransport.org/consortium/cons_faqs.cfm

On another point, I'm pretty sure my bios will allow me to change the bus
width between 8, 16, and 32bits wide. Now I'm wondering why anyone would
want to run it at half or quarter width. Or am I just not remembering
correctly. Hmmm... time to reboot.

--
Abit KT7-Raid (KT133) Tbred B core CPU @2400MHz (24x100FSB)
My server http://wesnewell.no-ip.com/cpu.php
Verizon server http://mysite.verizon.net/res0exft/cpu.htm

 

[Guest]

Archived from groups: alt.comp.hardware.overclocking,alt.comp.hardware.overclocking.amd (More info?)

On Tue, 05 Apr 2005 06:59:03 +0000, Wes Newell wrote:

> On another point, I'm pretty sure my bios will allow me to change the bus
> width between 8, 16, and 32bits wide. Now I'm wondering why anyone would
> want to run it at half or quarter width. Or am I just not remembering
> correctly. Hmmm... time to reboot.

Correction. It's 8, 16, and auto. Anyone know why this option whould be
useful?

--
Abit KT7-Raid (KT133) Tbred B core CPU @2400MHz (24x100FSB)
My server http://wesnewell.no-ip.com/cpu.php
Verizon server http://mysite.verizon.net/res0exft/cpu.htm

 

[Guest]

Archived from groups: alt.comp.hardware.overclocking,alt.comp.hardware.overclocking.amd (More info?)

"Michael Brown" <see@signature.below> wrote in message...
> Denny's actually somewhat right on this point.

And there's proof for the old truism about monkeys and the complete works of
Shakespeare. ;-)
--


Richard Hopkins
Cardiff, Wales, United Kingdom
(replace nospam with pipex in reply address)

The UK's leading technology reseller www.dabs.com

 

[Guest]

Archived from groups: alt.comp.hardware.overclocking,alt.comp.hardware.overclocking.amd (More info?)

Hi Wes:

Near as I can tell. The HT is a serial interface and the 8 or 16 bit you
are choosing is actually the bits per data transfer (same concept as DDR).
If you look at the data specifications in this first paragraph, it supports
that supposition...
http://www.hypertransport.org/tech/tech_specs.cfm

Here's a good read on the latest HT specifications...
http://www.hypertransport.org/tech/index.cfm

I can see no reason for having a user definable setting on the HT.

bye, Rick

"Wes Newell" <w.newell@TAKEOUTverizon.net> wrote in message
newsan.2005.04.05.07.27.47.402099@TAKEOUTverizon.net...
> On Tue, 05 Apr 2005 06:59:03 +0000, Wes Newell wrote:
>
>> On another point, I'm pretty sure my bios will allow me to change the bus
>> width between 8, 16, and 32bits wide. Now I'm wondering why anyone would
>> want to run it at half or quarter width. Or am I just not remembering
>> correctly. Hmmm... time to reboot.
>
> Correction. It's 8, 16, and auto. Anyone know why this option whould be
> useful?
>
> --
> Abit KT7-Raid (KT133) Tbred B core CPU @2400MHz (24x100FSB)
> My server http://wesnewell.no-ip.com/cpu.php
> Verizon server http://mysite.verizon.net/res0exft/cpu.htm
>

 

[Guest]

Archived from groups: alt.comp.hardware.overclocking,alt.comp.hardware.overclocking.amd (More info?)

Richard Savoie wrote:
> Hi Wes:
>
> Near as I can tell. The HT is a serial interface and the 8 or 16
> bit you are choosing is actually the bits per data transfer (same
> concept as DDR).

No. Section 2 of the HT spec.

"HyperTransport links wider than 8 bits are built by ganging multiple 8-bit
links in parallel to form either 16- or 32-bit links. Links wider than 8
bits have one clock per byte, but still only one CTL bit for the whole link.
The forwarded clock for the CTL signal is the clock for the
least-significant byte."

And in section 3:
"HyperTransport technology supports link widths of 2, 4, 8, 16, and 32
bits."

Translation: it's a parallel bus of variable width. In the case of the K8,
the maximum width in each direction is 16 bits, though it is configurable to
2, 4, 8, and 16 bits wide (the size in both directions is independent, so
you can have 2 bit downstream, 16 bit upstream if you wish). It does use
DDR, but that's pretty much irrelevant to the width of the bus.

[...]
> I can see no reason for having a user definable setting on the HT.

The only possible reason I can think of is if the motherboard was poorly
designed and had skew issues at high HT rates. Dropping down to a narrower
bus will reduce the maximum skew between any two lines. But if the board was
designed/tested properly in the first place then skew should not be much of
an issue.

[...]

--
Michael Brown
www.emboss.co.nz : OOS/RSI software and more
Add michael@ to emboss.co.nz ---+--- My inbox is always open

 

[Guest]

Archived from groups: alt.comp.hardware.overclocking,alt.comp.hardware.overclocking.amd (More info?)

"Wes Newell" <w.newell@TAKEOUTverizon.net> wrote in message
newsan.2005.04.05.07.01.24.441198@TAKEOUTverizon.net...
> On Tue, 05 Apr 2005 08:49:19 +1000, Michael Brown wrote:
>
> > Wes Newell wrote:
> >> On Mon, 04 Apr 2005 11:58:55 -0400, RaichuRaichu wrote:
> >>
> >>> A Computer shop owner I was talking to when I said about
> >>> AMD's FX55 @ 1,000 FSB said he had an Intel 775 (LGA)@
> >>> 2048 MHz FSB. 2.8GHz Or split to 4 for the 512 (True) FSB
> >>
> >> Just proves that computer shop owners can be ignorant too. The
> >> default FSB speed of the FX 55 is the same as all the other K8 cpu's,
> >> 200MHz.
> >
> > Denny's actually somewhat right on this point. The HT bus on an K8 does
run
> > at 800MHz or 1000MHz, and uses DDR on top of that (at 16 bits wide each
way
> > to get the 3.2 or 4 GBytes/sec bandwidth). The 200MHz is just used as a
> > common base clock for both the HT bus and the CPU speed.
I cant say as if I like TigerDirect, they are the cheapest for this chip.
http://www.tigerdirect.com/applications/SearchTools/item-details.asp?EdpNo=1
103939&CatId=1177

a total bus clock of 2,000 MHz
The RavingRaichu. ;-)
> >
> > [...]
>
> Yep, I guess I kind of blew that part.
>
> I've noticed that they've also raised to the allowable HT clock rate up to
> 1.4GHz since I last looked. I'm not sure why there's so much focus on
> this though (for the FSB). With a seperate memory controller, the A64
> doesn't need all of that bandwidth for other IO. For those interested;
>
> 14. At what clock speeds does HyperTransport technology operate?
> HyperTransport technology devices are designed to operate at multiple
> clock speeds from 200MHz up to 1.4 GHz, and utilize double data rate
> technology transferring two bits of data per clock cycle, for an effective
> transfer rate of up to 2.8 gigatransfer/sec in each direction. Since
> transfers can occur in both directions simultaneously, an aggregate
> transfer rate of 11.2 gigabytes per second in a 16 bit HyperTransport I/O
> Link and an aggregate transfer rate of 22.4 gigabytes per second in a
> 32-bit HyperTransport I/O Link can be achieved. To allow for system design
> optimization, the clocks of the receive and transmit links may beset at
> different rates.
>
> http://www.hypertransport.org/consortium/cons_faqs.cfm
>
> On another point, I'm pretty sure my bios will allow me to change the bus
> width between 8, 16, and 32bits wide. Now I'm wondering why anyone would
> want to run it at half or quarter width. Or am I just not remembering
> correctly. Hmmm... time to reboot.
>
> --
> Abit KT7-Raid (KT133) Tbred B core CPU @2400MHz (24x100FSB)
> My server http://wesnewell.no-ip.com/cpu.php
> Verizon server http://mysite.verizon.net/res0exft/cpu.htm
>