Performance Test Configuration
The six 2GB kits were tested with the DFI LANParty nF4 SLI-DR Athlon 64 Socket 939 motherboard. Other components remain the same as used in the memory setup in Athlon 64 Memory: Rewriting the Rules.
The A64 test bed includes components that have been proven in Socket 939 Athlon 64 benchmarking, such as the Socket 939 90nm 4000+, the OCZ Power Stream 520 Power Supply, and the NVIDIA 7800 GTX video card. We used an AMD 4000+ 90nm processor with a Revision E type memory controller. All other basic test conditions attempted to mirror those used in our earlier Athlon 64 memory reviews.
Since the video card, and platform/video drivers are upgraded from past memory testing, we found that the results were not completely consistent with past memory test results. For that reason, we retested the three 2GB kits that were previously tested. Results are still broadly comparable to past memory test results in earlier reviews, but we decided to include just the nine 2GB kit results in this roundup.
As discussed in other memory reviews, we ran a complete set of Memtest86 benchmarks with only tRAS varied to determine the best tRAS setting for these memories on the nForce4 chipset. The NVIDIA nForce4 has the best bandwidth at tRAS settings ranging from 5 to 8. Therefore, a tRAS setting of 7 was used for testing.
Test Settings
All AMD Athlon 64 processors are unlocked downward, and the FX CPUs are unlocked up and down. This feature allows a different approach to memory testing, which truly measures performance differences in memory speed alone. All tests were run with CPU speed as close to the specified 2.4GHz of the 4000+ as possible, with CPU speed/Memory Speed increased at lower multipliers to achieve 2.4 GHz. This approach allows the true measurement of the impact of higher memory speed and timings on performance, since CPU speed is fixed, removing CPU speed as a factor in memory performance.
The following settings were tested with the six 2GB kits on the DFI nF4 test bed:
The six 2GB kits were tested with the DFI LANParty nF4 SLI-DR Athlon 64 Socket 939 motherboard. Other components remain the same as used in the memory setup in Athlon 64 Memory: Rewriting the Rules.
The A64 test bed includes components that have been proven in Socket 939 Athlon 64 benchmarking, such as the Socket 939 90nm 4000+, the OCZ Power Stream 520 Power Supply, and the NVIDIA 7800 GTX video card. We used an AMD 4000+ 90nm processor with a Revision E type memory controller. All other basic test conditions attempted to mirror those used in our earlier Athlon 64 memory reviews.
Since the video card, and platform/video drivers are upgraded from past memory testing, we found that the results were not completely consistent with past memory test results. For that reason, we retested the three 2GB kits that were previously tested. Results are still broadly comparable to past memory test results in earlier reviews, but we decided to include just the nine 2GB kit results in this roundup.
| AMD nForce4 Performance Test Configuration | |
| Processor(s): | AMD 4000+ Athlon 64(90nm) 2.4GHz, Socket 939, 1 MB cache, Dual Channel, 1000HT |
| RAM: | TWINX2048-4000PT (DS) 2X1GB Crucial Ballistix CLIII5N.32 (DS) 2X1GB G. Skill F1-3200PHU2-2GBZX (DS) 2X1GB KHX3200AK2/2G (DS) 2X1GB Mushkin 2GB Redline XP4000 (DS) 2X1GB Team XTreem TXDR 1024M400HC2 (DS) 2X1GB Corsair TWINX2048-3500XL PRO (DS) 2X1GB Gigaram 2GB Dual Channel PC-4200 (DS) 2X1GB OCZ PC4000 2x1024MB EB Platinum (DS) 2X1GB |
| Hard Drives: | Seagate 120GB SATA 7200RPM 8MB Cache |
| PCI/AGP Speed: | Fixed at 33/66 |
| Bus Master Drivers: | NVIDIA nForce Platform Driver 6.70 |
| Video Card(s): | NVIDIA 7800 GTX 256MB PCIe, 256MB aperture, 1024x768x32 |
| Video Drivers: | NVIDIA Forceware 81.98 Release |
| Power Supply: | OCZ Power Stream 520W |
| Operating System(s): | Windows XP Professional SP2 |
| Motherboard: | DFI LANParty nF4 SLI-DR |
| BIOS: | 7/06/2005 |
As discussed in other memory reviews, we ran a complete set of Memtest86 benchmarks with only tRAS varied to determine the best tRAS setting for these memories on the nForce4 chipset. The NVIDIA nForce4 has the best bandwidth at tRAS settings ranging from 5 to 8. Therefore, a tRAS setting of 7 was used for testing.
Test Settings
All AMD Athlon 64 processors are unlocked downward, and the FX CPUs are unlocked up and down. This feature allows a different approach to memory testing, which truly measures performance differences in memory speed alone. All tests were run with CPU speed as close to the specified 2.4GHz of the 4000+ as possible, with CPU speed/Memory Speed increased at lower multipliers to achieve 2.4 GHz. This approach allows the true measurement of the impact of higher memory speed and timings on performance, since CPU speed is fixed, removing CPU speed as a factor in memory performance.
The following settings were tested with the six 2GB kits on the DFI nF4 test bed:
- 2.4GHz-12x200/DDR400 - the highest stock memory speed supported on nF3-4/SiS755-FX/ATI Radeon Xpress 200/VIA 939 motherboards
- 2.4GHz-11x218/DDR436 - a ratio near the standard DDR433 speed
- 2.4GHz-10x240/DDR480 - a ratio near the standard rating of DDR466
- 2.4GHz-9x267/DDR533 - a memory speed achieved by only a few top memories on the Athlon 64
- 2.4GHz-8x300/DDR600 - with recent improvements in OC abilities, a very few memories can reach this next multiple with the 8 multiplier
- Highest Memory Performance - the highest memory bandwidth and game performance that we could achieve with the memory being tested. This is the highest memory speed that we could achieve with 1T Command Rate.
Facebook
Twitter
RSS
51 Comments
View All Comments
DonPMitchell - Monday, March 6, 2006 - link
I'm surprised there was no discussion of single-rank vs. dual-rank memory. For the Athlon, that's a big issue, becuase its memory controller may down-clock beyond 4 ranks of memory.nyogen - Wednesday, February 15, 2006 - link
I remember when Nvidia released their 6xxx series there was an all out boycott of paper launches.OCZ and Mushkin released not very long ago their 2GB dual channel high end kits namely:
http://www.mushkin.com/doc/products/memory_detail....">http://www.mushkin.com/doc/products/memory_detail....
http://www.ocztechnology.com/products/memory/ocz_e...">http://www.ocztechnology.com/products/memory/ocz_e...
Apart from these parts not having been available, the new batch that carries only the name given by reviews are flagged by poor performance, different IC's from the original batch, and lots of glitches.
I am sure Corsair and all the other manufacturers are doing the same, paper launch, hard launch cherry picked and then selling out to unsuspecting customers a COMPLETELY different product, backed by reputable reviews.
Don't you feel like they have been using you and used your reputation and popularity in order to commit something very close to fraud ???
If you encourage us to buy these products then you have the responsability to follow up on the products life and batches so that once they try to do this again you will be there to sanction this and PROTECT US. Otherwise your reviews become fairy tales and the only ones that have the reviewed products are you and a couple of people in the world.
GO FOR THEM ! END THIS UNFAIR PRACTICE !
Overlag - Monday, February 6, 2006 - link
Crucial have infact stopped making there 1gig DDRI sticks due to the massive fail rate on there high end products.i have had 6x 512mb failures and 4x 1gig failures. While there customer services is the best of ANY online company ive dealt with, there high end products are really bad quality now.... the ram overheats, then after a few months wont run at 1t, then it starts making noises while you access it, before finaly degrading over a month from 2-2-2-6 settings down to 3-4-4-10 untill failing totaly.
This includes chips (my latest two) which ive been using at 2.7volts ONLY, and with active cooling, yet they are still hotter than my X2 4400 at 4800 speeds...Last week they run fine at 2-2-2-7 1t, now they are at 3-3-3-8 2t :(
... Why there high end products have problems i dont know, there normal or value ram runs fine for YEARS and YEARS.
Very disapointed with it all apart from the customer services. However even that has started to disapoint me, as they refused to RMA/refund me for my 2x 512mb pair...
leexgx - Monday, February 6, 2006 - link
i had no problems getting my ram RMAed and replaced (2 days ago 2 day turn around UK) never had something replaced so fast before :)you can still buy it from here (UK tho)
http://www.overclockers.co.uk/acatalog/Crucial_Dua...">http://www.overclockers.co.uk/acatalog/Crucial_Dua...
on an other note it is an consern to me as well i have never used ram that can heat my room (needs active cooling) like others have posted it does run hot if not cooled i cant seem to get to 2-2-2-7 T1 my asus a8n-sli prem does an anoying thing by auto setting it to 4 5 5 12 {fail safe maybe?}(when i open A64 tweeker as it shows the info) then the nvidia Sound or nvidia video tray Icon crashs the system with an nice BSOD, (works fine tho at 3-3-3-7 T1 @ ddr500 2.7v or ddr speed lower but just run it at ddr400 ish now)
an but realy any way good OCZ or just standered TwinMoss ram is realy real world 2-4% performance dif (Amd64 On chip ram contorler removes the bottle neck that was on the Older k7 chips i go for lower timeings over more bandwith)
realy should have an look at DFI streets Forums on the matter of ddr1 Ocing (AMD overclocking) http://www.dfi-street.com/forum/forumdisplay.php?f...">http://www.dfi-street.com/forum/forumdisplay.php?f...
(other note i think my ram was partly dead when i got it just put up with it untill me pc did not boot any more )
Overlag - Thursday, February 9, 2006 - link
i wouldnt want to buy ballistix ever again... Once this set finaly gives up the ghost i will ask for a full refund and get some real memory.XeCutor - Thursday, February 2, 2006 - link
My first pair failed memtest at 250mhz, seemed stable at 247mhz. My second pair does 250mhz but not much more. Fails at 255mhz. All this is on a DFI Expert board with an opteron 165 3-4-4-8 1T timings at 2,6-2,8v. Voltage doesn't seem to matter att all to these sticks, they fail at the same speeds regardless for me.bupkus - Wednesday, January 25, 2006 - link
I went for the Twinx2048-3200c2pt because of the lower timings and $197 AMIR price. I haven't researched this much lately, but I recall PC3200 with tight timings getting the nod from the enthusiast community for the Athlon 64.The TWINX2048-4000PT that I faced it off to lists timings of 3-4-4-8 which I thought wasn't preferable to the tighter timings especially with a price of $233.
The OCZ4002048ELDCPE-K is going for the same price of about $200 but the OCZ5001024EBPE only appears on newegg as a single 1GB module for $153.44 + $4.81 shipping.
That's too rich for my blood.
I guess the question for me would be which is the best for $200 +- $20?
Visual - Wednesday, January 25, 2006 - link
Besides the TWINX kit, corsair offers single 1GB modules too - again marked as 4000PTThe price for the kits for some reason seems higher than two separate modules.
So my question is, are the separate modules the same thing as what's in the kits? Why don't you mention them at all, and either recommend them for better value over the overpriced kits, or, if they dont perform as well, warn users about them?
Googer - Tuesday, January 24, 2006 - link
I would have liked to have seen four 512MB modules shown in the graphs as a refrence of how much or how little there is to gain by using these pairs of 1GB modules. By doing this I could form a cost/benefit analasys.bersl2 - Monday, January 23, 2006 - link
I'm going to wait to see if the price on DDR1 falls again before AMD goes DDR2. If it does, I'll get 2 x 1GiB to replace the 4 x 512MiB I bought over a year ago, but haven't been able to use until just now.