Next review (DDR3). How do you want it done?

[JN]

Here's the situation....

- We've got a P5K (DDR2) and a P5K3 (DDR3) motherboard

- We've got Kingston DDR2-1200 (5-5-5-15) and Kingston DDR3-1375 (7-7-7-20) 2gb memory kits

What kind of review would you like to see? It doesn't neccesarily need to include all of the above, but if it's a comparison, it needs to be as "fair" as possible.

Note: The DDR3 kit only runs up to 1360 on the P5K3.

I was thinking of doing a comparison review with the P5K running memory at 1066mhz (5-5-5-15) and the P5K3 running memory at 1333 (7-7-7-20), so they are at native supported speeds. This of course is providing the CPU can run at the same speed on both boards at these memory speeds.

What do you think?

 

[mVp24]

Here's the situation....

- We've got a P5K (DDR2) and a P5K3 (DDR3) motherboard

- We've got Kingston DDR2-1200 (5-5-5-15) and Kingston DDR3-1375 (7-7-7-20) 2gb memory kits

What kind of review would you like to see? It doesn't neccesarily need to include all of the above, but if it's a comparison, it needs to be as "fair" as possible.

Note: The DDR3 kit only runs up to 1360 on the P5K3.

I was thinking of doing a comparison review with the P5K running memory at 1066mhz (5-5-5-15) and the P5K3 running memory at 1333 (7-7-7-20), so they are at native supported speeds. This of course is providing the CPU can run at the same speed on both boards at these memory speeds.

What do you think?

Sounds good to me!

 

[PV5150]

I was thinking of doing a comparison review with the P5K running memory at 1066mhz (5-5-5-15) and the P5K3 running memory at 1333 (7-7-7-20), so they are at native supported speeds. This of course is providing the CPU can run at the same speed on both boards at these memory speeds.

Yep agreed. I'd like to see some of the usual slew of benchmarks that we run, plus a couple more if possible.

* RightMark Memory Analyzer

* SiSoft Sandra CPU, MMX, Memory Benchmarks

* Everest

* RealStorm

* HDTach

* WinRAR (data compression rate KB/s)

* TMPGEnc MPEG2 Encoding

* XviD Video Encoding

* LAME MP3 Encoding

* Gaming benchmarks

 

[ionicle]

can you not clock the DDR2 memory upto the same speed, i know it can clock that high, get it to the same latency too, then see what the difference is?

 

[JN]

can you not clock the DDR2 memory upto the same speed, i know it can clock that high, get it to the same latency too, then see what the difference is?

This could only really be done by loosening the DDR2 timings to the same as the DDR3....but then I have a feeling that there would be little to no difference between the two.

 

[Rastalovich]

That` s a tuffy.

How about some kind of % performance increase from stock to oc ? Even then tho it`s a bit of a mixed bag. But alteast the % will give a factor of stick v stick.

Best of luck with deciding.

 

[mVp24]

XMS, don't worry about really equaling their performance. The increase in speed is a special trait of DDR3, and if DDR2 OCs high, the DDR3 will go higher. Just run them with the highest OCs and tightest timings you can get. This will tell us if a lower speed and tighter timings of DDR2 are that much worse than higher speeds and loser timings of DDR3.

 

[jiffz]

Run the tests PV states at stock and then with both clocked and with tighter timings. Let users see the difference and the gain each set receives from overlocking

 

[Ham]

Why HDtach PVs?

Enless your wanting too see if theres a difference between the mobos, but theyre both P35 so the should be the same:eh:

I think you should bench @ native speeds, then a clock for clock with the DDR2 @ dd3 speed, then visa versa.

Thats alot of work though...

 

[ionicle]

howabout tightening the timings on the DDR3, and loosening the DDR2, so they're the average of both

then the same with the speed, increase the DDR2, by the same amount you decrease the DDR3, so they're both level

i want to know, when they're both at identical speeds, and timings, what the performance gain is, cos i dont see one tbh

 

[mrapoc]

Perhaps run the benchmark in "World in conflict" beta. You dont need the beta key to run the benchmark/tutorial. It p00ns my pc on very high settins (256+mb overload on card and below 20fps.)

 

[Mr. Smith]

I was thinking of doing a comparison review with the P5K running memory at 1066mhz (5-5-5-15) and the P5K3 running memory at 1333 (7-7-7-20), so they are at native supported speeds. This of course is providing the CPU can run at the same speed on both boards at these memory speeds.

What do you think?

[/SIZE]

Yep agreed. I'd like to see some of the usual slew of benchmarks that we run, plus a couple more if possible.

* RightMark Memory Analyzer

* SiSoft Sandra CPU, MMX, Memory Benchmarks

* Everest

* RealStorm

* HDTach

* WinRAR (data compression rate KB/s)

* TMPGEnc MPEG2 Encoding

* XviD Video Encoding

* LAME MP3 Encoding

* Gaming benchmarks

Seconded. There is no point tampering with the RAM. A top set of DDR2 vs a nice set of new DDR3. Basically old VS. new. See how the fair against eachother, what are the benefits,if any, of getting DDR3.

 

[JN]

Run the tests PV states at stock and then with both clocked and with tighter timings. Let users see the difference and the gain each set receives from overlocking

Ye unfortunately I cant even get the DDR3 to run at its stock speed so 'overclocking' is kinda out the question

Seconded. There is no point tampering with the RAM. A top set of DDR2 vs a nice set of new DDR3. Basically old VS. new. See how the fair against eachother, what are the benefits,if any, of getting DDR3.

1333 7-7-7-20 vs 1066 5-5-5-15 it is

 

[ionicle]

are they both the same quantity in Gb?

 

[FarFarAway]

are they both the same quantity in Gb?

Yes

 

[ionicle]

Yes

in that case, if the speed is the same, the timings are the same, and the quantity in Gb is the same, i'd like to see what the difference is

i know that DDR3 can run at a lower voltage though

but i'd like to see the actual difference ...

or is it just me?

its like DDR and DDR2, was there any difference, if they were both run at the same speed, same timings etc?