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

DDR2 Bandwidth Benchmarks: Post Here

Overclockers is supported by our readers. When you click a link to make a purchase, we may earn a commission. Learn More.
Sorry, I didn't run unbuffered. I won't have a chance to do that for awhile either. I'm in the middle of moving and have to be done by the end of the month. The extra computer stuff is already boxed up right now.
 
SiSoftware Sandra

Benchmark Results
RAM Bandwidth Int Buff'd iSSE2 : 6524 MB/s
RAM Bandwidth Float Buff'd iSSE2 : 6503 MB/s


Int Buff'd iSSE2 (Integer STREAM) Results Breakdown
Assignment : 6569MB/s
Scaling : 6307MB/s
Addition : 6711MB/s
Triad : 6510MB/s
Data Item Size : 16 byte(s)
Buffering Used : Yes
Offset Displacement Used : Yes
Bandwidth Efficiency : 73% (estimated)

Float Buff'd iSSE2 (Float STREAM) Results Breakdown
Assignment : 6468MB/s
Scaling : 6315MB/s
Addition : 6723MB/s
Triad : 6506MB/s
Data Item Size : 16 byte(s)
Buffering Used : Yes
Offset Displacement Used : Yes
Bandwidth Efficiency : 73% (estimated)

Chipset 1
Model : Abit Computer Corp ??? (2770)
Front Side Bus Speed : 4x 280MHz (1120MHz data rate)
Width : 64-bit
Maximum Bus Bandwidth : 8960MB/s (estimated)


This is at 3-3-3-11, 2.05v, FSB:Mem 1:1 at 280. I seem to get better numbers running at those speeds than at DDR700, 4-4-4-12.

Good info posted so far. Keep it up!! :)
 
Those are some amazing scores crimedog. I find it interesting that crimedog gets better scores at 1:1. I tried mine at 1:1 and my scores went down.

 
tried 5-3-3-8, and got lower scores than 3-3-3-11. :(

still playing.

Nice scores hyperasus. Insane scrores, actually. Especially at 293.
 
hyperasus said:
Those are some amazing scores crimedog. I find it interesting that crimedog gets better scores at 1:1. I tried mine at 1:1 and my scores went down.


well i got better scores 1:1 when the ram was around the same speed :p
 
crimedog would you be willing to try a couple of tests at or below 300fsb. I'm curious if your very high fsb has something to do with getting better scores at 1:1.
 
Pinky said:
It's not just how late the time is.

People in general are resistent to change. It's also hard to convince someone who just spent $200 on DDR that DDR2 is decent stuff. Add to that the fact DDR2 will be the standard beginning next year (rendering their recent $200 purchase obsolete in future motherboards/upgrade routes). You could say it's partially resentment towards an ever-changing industry. Staying on top could break the bank. It does mine! lol Oh, and we're mostly geeks. We take technology a tad too seriously by nature. PC components are part of the family.

I've been around here for a few days ;), and have seen a fair number of changes. :p
I agree. If there was a fire I would get the kids out then go back for the PC's.... Or maybe have each take a PC with them.. Hmmm.

The wife, BAH! She is on her own. :santa:
 
try doing the math like it says and comparing it to the latencies you already see as test examples. youll see that it gives a clear picture which ram is superior in terms of latency.

200mhz~5ns ... 2+2+2+5=11 ... 5x11 = 55.ns

280mhz~3.6ns ... 2+2+2+5= 11 ... 3.6x11 = 39.6ns

360mhz~2.8ns ... 3+2+2+4= 11 ... 2.8x11 = 30.8ns

looks about right to me. but im by no means perfect feel free to correct me with the right answer.


i could have misunderstood thier explanation of CL and AL maybe its only the cas "CL" that you measure. and the rest may stand for one or more extra "AL" as you said it not completly simple, but anything is simple to someone who understands it.

we just need to find out the right answer

EDIT: i misunderstood here is the correct info

heres the link to the info

http://www.hardwaresecrets.com/article/167/4


to highlight the important parts
as we all seem to know Cas (the first timing number) is by far the most important in terms of memory performance. as highlighted here


"If we compare a DDR memory to a DDR2 memory running under the same clock, the one with the lower latency will be the fastest. Thus, if you have a DDR400 CL3 memory and a DDR2-400 CL4 memory, your DDR400 will be faster."


however what isnt common knowledge is this


"Keep in mind that DDR2 memories have an additional parameter called AL (additional latency), which must be added to their nominal latency (CL) in order to get the total latency."


so this means CAS + AL for DDR2 vs just CAS for DDR


"The additional latency (AL) is usually found on the memory module technical specs (usually a PDF file for downloading on the manufacturer website)."


now the breakdown of MHZ and how it works with the timings.


"On a DDR400 CL3 memory, this “3” means that the memory delays three clock cycles to start delivering the requested data. Since this memory runs at 200 MHz, each clock tick measures 5 ns (T= 1/f). Thus its latency if of 15 ns.

Now on a DDR2-533 CL3 AL0 memory, this “3” also means that the memory delays three clock cycles to start delivering the request data, but since this memory runs at 266 MHz, each clock tick measures 3.75 ns, so its latency is of 11.25 ns, making this memory faster to data delivery than our DDR400 CL3 memory. So a DDR2-533 CL4 and AL0 memory has the same latency as a DDR400 CL3. Notice that we are assuming the additional latency as zero, or we would need to take it into account, i.e., a DDR2 CL3 AL1 memory has in reality a latency of four clock cycles which would equal 15ns."


the clock tick duration for memory speeds

Memory.........................Clock Tick Duration (each one)
DDR266.........................7.5 ns
DDR333.........................6 ns
DDR400 and DDR2-400.....5 ns
DDR2-533......................3.75 ns
DDR2-667......................3 ns
DDR2-800......................2.5 ns






dont get AL (additional latency) mixed up with the other timings. the other timings mean this

* CL: CAS Latency. The time it takes between a command having been sent to the memory and when it begins to reply to it. It is the time it takes between the processor asking for some data from the memory and it returning it.
* tRCD: RAS to CAS Delay. The time it takes between the activation of the line (RAS) and the column (CAS) where the data are stored in the matrix.
* tRP: RAS Precharge. The time it takes between disabling the access to a line of data and the begin of the access the another line of data.
* tRAS: Active to Precharge Delay. How long the memory has to wait until the next access to the memory can be initiated.
* CMD: Command Rate. The time it takes between the memory chip having been activated and when the first command may be sent to the memory. Sometimes this value is not informed. It usually is T1 (1 clock speed) or T2 (2 clock speeds).
 
Last edited:
Guys... I feel I need help/guidance here.

Firstly, I'm running 2 x 512 sticks of Ballistic 667. I have manually set the voltage to 2.15 in BIOS and left all other memory settings alone. I am running a 660cpu @ 290x14 at present, and looking at your results... suspect mine are somewhat on the low side... I have pasted most of what I get from Sandra, hoping that might assist.

Appreciate any help

B

RAM Bandwidth Int Buff'd iSSE2 : 7017 MB/s
RAM Bandwidth Float Buff'd iSSE2 : 7013 MB/s
Results Interpretation : Higher index values are better.

Int Buff'd iSSE2 (Integer STREAM) Results Breakdown
Assignment : 6813MB/s
Scaling : 6822MB/s
Addition : 7214MB/s
Triad : 7220MB/s
Data Item Size : 16 byte(s)
Buffering Used : Yes
Offset Displacement Used : Yes
Bandwidth Efficiency : 76% (estimated)

Float Buff'd iSSE2 (Float STREAM) Results Breakdown
Assignment : 6816MB/s
Scaling : 6813MB/s
Addition : 7211MB/s
Triad : 7212MB/s
Data Item Size : 16 byte(s)
Buffering Used : Yes
Offset Displacement Used : Yes
Bandwidth Efficiency : 76% (estimated)

Performance Test Status
Run ID : OASIS on Thursday, 3 November 2005 at 12:59:12 p.m.
Memory Used by Test : 512MB
NUMA Support : No
SMP Test : No
Total Test Threads : 1
Multi-Core Test : No
SMT Test : Yes
Dynamic MP/MT Load Balance : No
Processor Affinity : P0C0T0
Page Size : 4kB
Use Large Memory Pages : No

Chipset 1
Model : ASUSTeK Computer Inc 82955x Processor to I/O Controller
Front Side Bus Speed : 4x 290MHz (1160MHz data rate)
Width : 64-bit
Maximum Bus Bandwidth : 9280MB/s (estimated)

Logical/Chipset 1 Memory Banks
Bank 0 : 256MB DDR2-SDRAM 5.0-6-6-2 1CMD
Bank 1 : 256MB DDR2-SDRAM 5.0-6-6-2 1CMD
Bank 4 : 256MB DDR2-SDRAM 5.0-6-6-2 1CMD
Bank 5 : 256MB DDR2-SDRAM 5.0-6-6-2 1CMD
Channels : 2
Speed : 4x 193MHz (772MHz data rate)
Width : 64-bit
Performance Acceleration Technology : Yes
Maximum Memory Bus Bandwidth : 12352MB/s (estimated)
 
Oh... okay. I thought I should have been somewhere around the 8000 mark rather than 7000ish.

Runs fine, so happy enough, I just wondered really if I should have been using a different multiplier (still trying to work DDR2 out really) and whether or not I should be playing with timings or just leave them all at auto?
 
your timings look a little weird to me.

Bank 0 : 256MB DDR2-SDRAM 5.0-6-6-2 1CMD
Bank 1 : 256MB DDR2-SDRAM 5.0-6-6-2 1CMD
Bank 4 : 256MB DDR2-SDRAM 5.0-6-6-2 1CMD
Bank 5 : 256MB DDR2-SDRAM 5.0-6-6-2 1CMD

I would look into trying to get better latencies. 6 is rather high. You will get better efficiency and better scores by tightening the timings, if possible.
 
Okay... went to manual and it had these in there which only produced a slight gain...

RAM Bandwidth Int Buff'd iSSE2 : 7188 MB/s
RAM Bandwidth Float Buff'd iSSE2 : 7194 MB/s
Results Interpretation : Higher index values are better.

Int Buff'd iSSE2 (Integer STREAM) Results Breakdown
Assignment : 6967MB/s
Scaling : 6957MB/s
Addition : 7416MB/s
Triad : 7415MB/s
Data Item Size : 16 byte(s)
Buffering Used : Yes
Offset Displacement Used : Yes
Bandwidth Efficiency : 77% (estimated)

Float Buff'd iSSE2 (Float STREAM) Results Breakdown
Assignment : 6986MB/s
Scaling : 6967MB/s
Addition : 7412MB/s
Triad : 7411MB/s
Data Item Size : 16 byte(s)
Buffering Used : Yes
Offset Displacement Used : Yes
Bandwidth Efficiency : 78% (estimated)

Performance Test Status
Run ID : OASIS on Friday, 4 November 2005 at 8:33:59 a.m.
Memory Used by Test : 512MB
NUMA Support : No
SMP Test : No
Total Test Threads : 1
Multi-Core Test : No
SMT Test : Yes
Dynamic MP/MT Load Balance : No
Processor Affinity : P0C0T0
Page Size : 4kB
Use Large Memory Pages : No

Chipset 1
Model : ASUSTeK Computer Inc 82955x Processor to I/O Controller
Front Side Bus Speed : 4x 290MHz (1160MHz data rate)
Width : 64-bit
Maximum Bus Bandwidth : 9280MB/s (estimated)

Logical/Chipset 1 Memory Banks
Bank 0 : 256MB DDR2-SDRAM 5.0-4-4-15 1CMD
Bank 1 : 256MB DDR2-SDRAM 5.0-4-4-15 1CMD
Bank 4 : 256MB DDR2-SDRAM 5.0-4-4-15 1CMD
Bank 5 : 256MB DDR2-SDRAM 5.0-4-4-15 1CMD
Channels : 2
Speed : 4x 193MHz (772MHz data rate)
Width : 64-bit
Performance Acceleration Technology : Yes
Maximum Memory Bus Bandwidth : 12352MB/s (estimated)

Features
(W)MMX Technology : Yes
SSE Technology : Yes
SSE2 Technology : Yes
EMMX - Extended MMX Technology : No
HTT - Hyper-Threading Technology : Yes
 
No offense intel users, but I'd really like to see what happens w/ AMDs on DDR2, because of their memory controller I'd assume you'd get much better speeds, but I dunno. Just thought I'd put that out there...not trying to start anything.
 
Back