DDR SDRAM Versus Direct RDRAM @ Ace's Hardware

MadRat

Lifer
Oct 14, 1999
11,999
307
126
Just some initial criticism:

1. What, no 333DDR? ...No 400DDR?
2. The charts are messy - its hard to follow at times
3. They only benchmark 2.0GHz-P4, 2.0GHz-P4A and 2.2GHz P4's. No 1.6A/1.8A@2.2GHz+? :(

It is nice to see the SDRAM vs. DDR RAM vs. PC800 vs PC1066 at 2.0GHz. It was hard to tell the difference between RDRAM (what they called DRDRAM) and DDR RAM by the charts. They should have used labels like PC133, PC1600, PC2100, PC800, PC1066, etc. They should have also kept all three like systems alongside the others rather than mixing them all together. Throwing in AMD benchmarks to prove the differences in memory performance also made it cluttered. The article was just plain hard to read and make sensical conclusions at first glance.
 

THUGSROOK

Elite Member
Feb 3, 2001
11,847
0
0
obviously they did alot of work but...

that info is completly worthless to us P4A overclockers :(
 

Athlon4all

Diamond Member
Jun 18, 2001
5,416
0
76
repost;)

<< What, no 333DDR? >>

There was DDR333 on SiS 645, you can't run DDR333 on Asus P4B266 without having the fsb below 533fsb.

<< No 400DDR? >>

At the end of the article, he gives several very good reasons showing that DDR is on it's last legs and I personally don't think DDR400 will ever hit the market so I don't think it should even be considered.

<< They only benchmark 2.0GHz-P4, 2.0GHz-P4A and 2.2GHz P4's No 1.6A/1.8A@2.2GHz+? >>

Well, Johan doesn't have the same resources that reviewers like Anand has and plus, he has been working on this for a long time (I emailed him about something and he mentioned this artilce over a month ago). Besides, he did test 533fsb with the 2GHz P4 and he did benchmark a 2.2 P4A, and the purpose of this article was really comparing all the different platforms and leading boards and memory types of today and of the future.

EDIT:

<< that info is completly worthless to us P4A overclockers >>

How is it useless? It will show people that 850 is clearly the supirior platform and DDR shouldn't even be considered with current prices.
 

manko

Golden Member
May 27, 2001
1,846
1
0
an interesting minor point was that he found he could only run his DDR at 2.5 on the SiS 645 boards
 

Athlon4all

Diamond Member
Jun 18, 2001
5,416
0
76


<< an interesting minor point was that he found he could only run his DDR at 2.5 on the SiS 645 boards >>

That's just another example of showing that DDR is having trouble at DDR333 speeds.
 

THUGSROOK

Elite Member
Feb 3, 2001
11,847
0
0
alot of ppl wont even concider gettin into bed with RDram.

the future? no dual channel ddr tests? the future is unknown - how about testing for today?

those tests have been done before - we were hoping for something a bit more insightful....
like a 1.6A @ 2500 or a 1.8A @ 2400 vs a stock 2.4A, etc.

(but chances are they wouldnt have used ddr333 anyways) :(
 

JellyBaby

Diamond Member
Apr 21, 2000
9,159
1
81
Yes the reviewer seemed quite enthralled by rdram. To me it seemed DDR stood neck-in-neck with it. At current clock speeds DDR looks perfectly fine. Tomorrow, who knows but tomorrow we may have DDR II and QDM.

I would also have like to see some extreme 1.xA overclocking results.
 

Athlon4all

Diamond Member
Jun 18, 2001
5,416
0
76
DDR may be sufficent now, but once 850e hits with 533fsb, RDRAM will take the lead and won't fall back until DDR-II, and even when Dual Channel DDR does hit, it will likely be so expensive that RDRAM will be the better option.
 

AGodspeed

Diamond Member
Jul 26, 2001
3,353
0
0
Has anyone got any solid confirmation of i845 (Grantie Bay) Dual Channel DDR chipsets in the pipeline for desktop users?

Looking at Johan's review, it's clear that the ASUS P4T-E (i850) is one stable, high-performing, and overclockable mofo. It's clearly THE P4 board to get (unless the feature-rich Soyo board is your kind of thing). Johan did have some problems with the Soyo board, but it looked like it was just with overclocking.

And it pains me to mention this VIA quote from Johan, as it's starting to get annoying:

Many KT266, P4X266, and KT266A tend to lock up sporadically during intensive rendering...

He does go on to mention that it depends on what mainboard maker is implementing the VIA chipset in question, and in that case Tyan did a good job implementing VIA's P4X266 chipset. Ummmmmmmm