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 » Motherboards » Asus Motherboards
Site Map Home Register Authors List Search Today's Posts Mark Forums Read Web Partners

fsb:dram & p5l-mx



 
 
Thread Tools Display Modes
  #1  
Old December 13th 07, 01:17 PM posted to alt.comp.periphs.mainboard.asus
Trin
external usenet poster
 
Posts: 4
Default fsb:dram & p5l-mx

Hi,

I have a p5l-mx mobo and i've been researching something
for half a day.

I'm using core 2 duo 4300 at 800 FSB. My memory is
DDR2-667 with a fsb:dram ratio of 3:5. It is said that
1:1 fsb:dram ratio is optimum. Does this mean using a
DDR400 giving a ratio of 1:1 would produce faster output
than the DDR2-667? If not, why not?

Thanks.

Tri
  #2  
Old December 13th 07, 01:50 PM posted to alt.comp.periphs.mainboard.asus
Paul
external usenet poster
 
Posts: 13,364
Default fsb:dram & p5l-mx

Trin wrote:
Hi,

I have a p5l-mx mobo and i've been researching something
for half a day.

I'm using core 2 duo 4300 at 800 FSB. My memory is
DDR2-667 with a fsb:dram ratio of 3:5. It is said that
1:1 fsb:dram ratio is optimum. Does this mean using a
DDR400 giving a ratio of 1:1 would produce faster output
than the DDR2-667? If not, why not?

Thanks.

Tri


The faster it goes, the better. I think it is possible
the FSB is FSB1066 here, so DDR2-533 is a good starting
point, but the faster RAM options do help.

http://www.anandtech.com/showdoc.aspx?i=2863&p=5

It is all a question, of whether the percentage increase
in application performance, justifies spending more for
the memory.

Paul
  #3  
Old December 13th 07, 02:52 PM posted to alt.comp.periphs.mainboard.asus
Trin
external usenet poster
 
Posts: 4
Default fsb:dram & p5l-mx

On Dec 13, 9:50 pm, Paul wrote:
Trin wrote:
Hi,


I have a p5l-mx mobo and i've been researching something
for half a day.


I'm using core 2 duo 4300 at 800 FSB. My memory is
DDR2-667 with a fsb:dram ratio of 3:5. It is said that
1:1 fsb:dram ratio is optimum. Does this mean using a
DDR400 giving a ratio of 1:1 would produce faster output
than the DDR2-667? If not, why not?


Thanks.


Tri


The faster it goes, the better. I think it is possible
the FSB is FSB1066 here, so DDR2-533 is a good starting
point, but the faster RAM options do help.


The P5l-MX has maximum of FSB 800 & DDR2-667. I remember
reading a year back or so that DDR2-667 is slower than DDR2-533
or DDR2-400 and I can't find the articles anymore. What
configuration can this slowing down occur?

Tri

http://www.anandtech.com/showdoc.aspx?i=2863&p=5

It is all a question, of whether the percentage increase
in application performance, justifies spending more for
the memory.

Paul


  #4  
Old December 13th 07, 03:36 PM posted to alt.comp.periphs.mainboard.asus
Paul
external usenet poster
 
Posts: 13,364
Default fsb:dram & p5l-mx

Trin wrote:
On Dec 13, 9:50 pm, Paul wrote:
Trin wrote:
Hi,
I have a p5l-mx mobo and i've been researching something
for half a day.
I'm using core 2 duo 4300 at 800 FSB. My memory is
DDR2-667 with a fsb:dram ratio of 3:5. It is said that
1:1 fsb:dram ratio is optimum. Does this mean using a
DDR400 giving a ratio of 1:1 would produce faster output
than the DDR2-667? If not, why not?
Thanks.
Tri

The faster it goes, the better. I think it is possible
the FSB is FSB1066 here, so DDR2-533 is a good starting
point, but the faster RAM options do help.


The P5l-MX has maximum of FSB 800 & DDR2-667. I remember
reading a year back or so that DDR2-667 is slower than DDR2-533
or DDR2-400 and I can't find the articles anymore. What
configuration can this slowing down occur?

Tri

http://www.anandtech.com/showdoc.aspx?i=2863&p=5

It is all a question, of whether the percentage increase
in application performance, justifies spending more for
the memory.

Paul



In previous generations of computers, there was a difference
between "synchronous" operation, and async operation. When running
async, resynchronization would add latency (delay) to the path between
memory and processor. The result was, that to get an advantage
from running async, you had to crank the memory quite high,
to compensate for the extra latency.

To give an example, if you had a FSB266 processor, and tried to
run memory at DDR333, it might be slower than if run at DDR266.
That is because the DDR333 needed resynchronization. If the
memory speed was further increased to DDR400, then that config
of FSB266/DDR400, would be faster than FSB266/DDR266.

That no longer seems to be the case. Now you can run async, and
have memory faster than FSB, and get some additional benefit.

Paul
  #5  
Old December 13th 07, 10:09 PM posted to alt.comp.periphs.mainboard.asus
Trin
external usenet poster
 
Posts: 4
Default fsb:dram & p5l-mx

On Dec 13, 11:36 pm, Paul wrote:
Trin wrote:
On Dec 13, 9:50 pm, Paul wrote:
Trin wrote:
Hi,
I have a p5l-mx mobo and i've been researching something
for half a day.
I'm using core 2 duo 4300 at 800 FSB. My memory is
DDR2-667 with a fsb:dram ratio of 3:5. It is said that
1:1 fsb:dram ratio is optimum. Does this mean using a
DDR400 giving a ratio of 1:1 would produce faster output
than the DDR2-667? If not, why not?
Thanks.
Tri
The faster it goes, the better. I think it is possible
the FSB is FSB1066 here, so DDR2-533 is a good starting
point, but the faster RAM options do help.


The P5l-MX has maximum of FSB 800 & DDR2-667. I remember
reading a year back or so that DDR2-667 is slower than DDR2-533
or DDR2-400 and I can't find the articles anymore. What
configuration can this slowing down occur?


Tri


http://www.anandtech.com/showdoc.aspx?i=2863&p=5


It is all a question, of whether the percentage increase
in application performance, justifies spending more for
the memory.


Paul


In previous generations of computers, there was a difference
between "synchronous" operation, and async operation. When running
async, resynchronization would add latency (delay) to the path between
memory and processor. The result was, that to get an advantage
from running async, you had to crank the memory quite high,
to compensate for the extra latency.

To give an example, if you had a FSB266 processor, and tried to
run memory at DDR333, it might be slower than if run at DDR266.
That is because the DDR333 needed resynchronization. If the
memory speed was further increased to DDR400, then that config
of FSB266/DDR400, would be faster than FSB266/DDR266.


How come in the above example the DDR400 doesn't need
resynchronization making the FSB266/DDR400 combo as you
mentioned faster?


That no longer seems to be the case. Now you can run async, and
have memory faster than FSB, and get some additional benefit.

Paul- Hide quoted text -

- Show quoted text -


How do you tell which mobo has the invariant syncs
specs that doesn't produce slowdown compared to
those that produce slowdowns? Any particular bios
signature or something? Or northbridge feature? I
heard core 2 duos needs synchronization for faster
results? It is no longer required too in newer mobos?


Trin
  #6  
Old December 13th 07, 10:52 PM posted to alt.comp.periphs.mainboard.asus
DaveW[_5_]
external usenet poster
 
Posts: 220
Default fsb:dram & p5l-mx

Wrong. You are confusing DDR RAM and the newer and faster DDR2 RAM. You
need to use ONLY DDR2 RAM in that motherboard. It is incompatible with the
older DDR RAM.


--
--DaveW
"Trin" wrote in message
...
Hi,

I have a p5l-mx mobo and i've been researching something
for half a day.

I'm using core 2 duo 4300 at 800 FSB. My memory is
DDR2-667 with a fsb:dram ratio of 3:5. It is said that
1:1 fsb:dram ratio is optimum. Does this mean using a
DDR400 giving a ratio of 1:1 would produce faster output
than the DDR2-667? If not, why not?

Thanks.

Tri



  #7  
Old December 13th 07, 11:00 PM posted to alt.comp.periphs.mainboard.asus
Trin
external usenet poster
 
Posts: 4
Default fsb:dram & p5l-mx

On Dec 14, 6:52 am, "DaveW" wrote:
Wrong. You are confusing DDR RAM and the newer and faster DDR2 RAM. You
need to use ONLY DDR2 RAM in that motherboard. It is incompatible with the
older DDR RAM.


I know since the PL5-MX only allows DDR2-400, DDR2-533 and
DDR2-667. But since the fsb of the mainboard is 200 x 4 = 800.
The memory should supposedly be DDR2-400 so the fsb:dram
can be 1:1 due to the similar 200 initial clock. What I'd like
to know is how does one tell which mainboard doesn't cause
slowdown from resynchronization? What particular technical
spec to look for and how does those newer boards able to make
it in such a way that no resynchronization is required in say
using a DDR2-667 to produce an fsb:dram of 3:5 from the memory
333 clock versus the 200 clock of the cpu?

Tri

--
--DaveW"Trin" wrote in message

...



Hi,


I have a p5l-mx mobo and i've been researching something
for half a day.


I'm using core 2 duo 4300 at 800 FSB. My memory is
DDR2-667 with a fsb:dram ratio of 3:5. It is said that
1:1 fsb:dram ratio is optimum. Does this mean using a
DDR400 giving a ratio of 1:1 would produce faster output
than the DDR2-667? If not, why not?


Thanks.


Tri- Hide quoted text -


- Show quoted text -


  #8  
Old December 14th 07, 04:53 AM posted to alt.comp.periphs.mainboard.asus
Paul
external usenet poster
 
Posts: 13,364
Default fsb:dram & p5l-mx

Trin wrote:


To give an example, if you had a FSB266 processor, and tried to
run memory at DDR333, it might be slower than if run at DDR266.
That is because the DDR333 needed resynchronization. If the
memory speed was further increased to DDR400, then that config
of FSB266/DDR400, would be faster than FSB266/DDR266.


How come in the above example the DDR400 doesn't need
resynchronization making the FSB266/DDR400 combo as you
mentioned faster?


That is because there are two factors at work. The latency introduced
by resynchronization is one effect. It is a negative. But when the
memory gets faster, that is a positive. At DDR400, the positive factor
has a greater effect, than the negative one, and the configuration is
better as a result.


That no longer seems to be the case. Now you can run async, and
have memory faster than FSB, and get some additional benefit.

Paul- Hide quoted text -

- Show quoted text -


How do you tell which mobo has the invariant syncs
specs that doesn't produce slowdown compared to
those that produce slowdowns? Any particular bios
signature or something? Or northbridge feature? I
heard core 2 duos needs synchronization for faster
results? It is no longer required too in newer mobos?


Trin


I haven't seen any performance results for DDR2, that differ
from the one I showed. So I still think you will be ahead by
using a faster RAM. But the improvement is percentage points,
so you can look at the cost, and the amount of improvement,
and decide whether it is worthwhile or not.

Running in dual channel, with a couple matched sticks, is
going to help you as much as some difference in the clock
rate. So at least use a dual channel setup.

Paul
  #9  
Old December 14th 07, 04:57 AM posted to alt.comp.periphs.mainboard.asus
Paul
external usenet poster
 
Posts: 13,364
Default fsb:dram & p5l-mx

Paul wrote:
Trin wrote:


To give an example, if you had a FSB266 processor, and tried to
run memory at DDR333, it might be slower than if run at DDR266.
That is because the DDR333 needed resynchronization. If the
memory speed was further increased to DDR400, then that config
of FSB266/DDR400, would be faster than FSB266/DDR266.


How come in the above example the DDR400 doesn't need
resynchronization making the FSB266/DDR400 combo as you
mentioned faster?


That is because there are two factors at work. The latency introduced
by resynchronization is one effect. It is a negative. But when the
memory gets faster, that is a positive. At DDR400, the positive factor
has a greater effect, than the negative one, and the configuration is
better as a result.


That no longer seems to be the case. Now you can run async, and
have memory faster than FSB, and get some additional benefit.

Paul- Hide quoted text -

- Show quoted text -


How do you tell which mobo has the invariant syncs
specs that doesn't produce slowdown compared to
those that produce slowdowns? Any particular bios
signature or something? Or northbridge feature? I
heard core 2 duos needs synchronization for faster
results? It is no longer required too in newer mobos?


Trin


I haven't seen any performance results for DDR2, that differ
from the one I showed. So I still think you will be ahead by
using a faster RAM. But the improvement is percentage points,
so you can look at the cost, and the amount of improvement,
and decide whether it is worthwhile or not.

Running in dual channel, with a couple matched sticks, is
going to help you as much as some difference in the clock
rate. So at least use a dual channel setup.

Paul


Also, it occurs to me, that you can buy DDR2-667 memory, and
experiment with it. You can change the memory bus clock and test
performance by using an application. Use something like SuperPI
for example, and see which speed setting is best. Buying the
faster RAM, will allow you to run at all three supported speeds.
Faster RAM can run at a slower speed, if desired.

http://www.xtremesystems.com/pi/
http://www.xtremesystems.com/pi/super_pi_mod-1.5.zip

Paul
 




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
ZIP DRAM? Pelysma General 1 June 22nd 05 07:31 AM
Now there's DRAM remarking YKhan General 0 January 17th 05 06:09 PM
The future of DRAM Radith General 0 June 25th 04 07:10 AM
Documentation about DRAM outsider General 0 January 16th 04 01:32 PM
They're Baa-aack........ with XDR DRAM George Macdonald General 56 July 30th 03 02:05 PM


All times are GMT +1. The time now is 11:36 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.