• Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!

Dual DDR really 2 x faster?

Overclockers is supported by our readers. When you click a link to make a purchase, we may earn a commission. Learn More.

Rave

Registered
Joined
Nov 5, 2002
I have posted this as a reponse to another thread but in doing so I am really interested what you guys think? So sorry for doubling up.

Well from what I have seen of the dual DDR it does not actually provide 2X the theoritcal bandwidth at all. This is the review that Tom's Hardware did on the 2800+ Athlon using Dual DDR 333 nforce2.

From the Sisoft Sandra benchmarks I can not see any improvement from the dual DDR as the p4 with RD still beats it easily. Oh and by the way so does my P4 with DDR 400 beats the athlon 2800+ easily with nforce2 I get about 3000 in this benchy.

Half way between the Athlon using nforce2 dual ddr and rd P4.

I am very confused by this as it does not appear that this is any extra what so ever? Can anyone explain this clearly for me?

Here you can see what I am talking about... http://www6.tomshardware.com/cpu/02...xp_2800-17.html

And here is the whole article ... http://www.tomshardware.com/cpu/02q4/021001/index.html
 

imgod2u

Member
Joined
Jun 8, 2002
Location
Isla Vista, CA
The Athlon's FSB is 166MHz DDR single channel. That dual channel DDR at 166MHz DDR isn't going to be much help and the memory benchmarks are based on CPU to memory bandwidth so it won't improve much with dual channel DDR either. Just wait for Dual Channel DDR to be used with a processor with a fast enough FSB to saturate the memory bandwidth (i.e. the P4). I think you'll be quite pleased with the memory bandwidth improvement.
 
OP
R

Rave

Registered
Joined
Nov 5, 2002
So atm the athlon is a bag of **** still with or without the dual ddr. Unless of course you are comparing theoritical cpu's that may or may not be available early next year...sounds like the Nvidia clutch at straws...

but thanx dood for clearing that up... good to know the p4 reigns supreme regardless of the nforce2 dual ddr at the moment!
 

Buzzdog

Member
Joined
Jul 11, 2002
Location
BFE, IN
Rave,

I would consider the source of that article also. On more then one occasion there has been questionable reviews from that source. I would agree with imgod2u, you will not see as big of a performance jump with the AMD chips as you would with a PIV using DCDDR. The PIV's have a bigger pipe to the memory then the XP's do. When DCDDR hits the PIV you should see performance close to what you see out of RDRAM. Just remember that we will no longer be using 3:4 memory dividers with DCDDR as this would create a bottle neck at the memory bus. That is why the AMD's are not showing well with DCDDR, they are choking with a bottle neck at the CPU.


Buzzdog
 

Buzzdog

Member
Joined
Jul 11, 2002
Location
BFE, IN
For one I can remember a picture he had posted of a PIV that was quickly removed when people noticed that it was an old stock photo of a 3.06 or something. It is nothing personal it just seems that he is brand biased. I have found that the information that is given here at overclockers.com is for the most part unbiased and to the point. I made 1 purchase decission based on tomshardware, my gigabyte board. The motherboard has been stable for me, but it is not the best RDRAM board to overclock with IMHO. I just wish I would have waited a little bit and found this sight before I had made that purchase. You can find other threads here that are critical of that sight. Dont get me wrong I am sure there is usefull information there. The article I am thinking of was about 2 months ago maybe. I remember there was alot said here on the forums regarding it. I cant remember all of the details persay just that he claimed to have a high speed PIV and the picture did not even match. Shortly after the picture was removed from the article. I am sure someone else in the forums here will have more information about it.

Buzzdog
 

Tipycol

Member
Joined
Aug 20, 2002
imgod2u said:
Anand's review of the nForce 2 showed little improvement over the KT333 solution as well except in the memory benchmarks.

You should check their site again, they made another review about the chipset. I'm not saying it's worth buying if you have a good board already, it just might be a good choice for someone that's upgrading their entire system (or building a new one)

LINK
 

@[email protected]

Senior Member
Joined
Dec 18, 2000
Location
Chapecó-SC
Neo_peter said:
what's wrong with tom's hardware specificically? which articles were questionable?

That´s not the kind of info we save in our brain HD...:p But I´ve listened a lot about it... If you want a very critic, unbiased information about what´s going on, check the Ed Stroglio reviews in the front page. He always read all the sites info and reviews and make a coment about all them showing what is what and what should be and all the possible "why" things... That info and the less biased people are in OC.com, my opinion offcourse...
 

@[email protected]

Senior Member
Joined
Dec 18, 2000
Location
Chapecó-SC
If I´m not mistunderstood, they published an old revision about the first celeron pIV based, that shi* with 128k L2 wich they insisted to show as a good buy...
 

quegyboe

Member
Joined
Dec 20, 2000
Location
BC Canada
On the clawhammer, I don't really see any diff between Dual Channel and regular DDR setups. As for the Sledgehammer, I can see it making a diff there because the Sledge supports it. Also, the Pentium 4 can take advantage of it, because of it's bandwidth hungry design.