Aren't these exactly the same? Why do they cost $50 apart??

EvilLost

Distinguished
Sep 1, 2006
31
0
18,530

EvilLost

Distinguished
Sep 1, 2006
31
0
18,530
looking at their "naming specs" it seems that the cheaper one is "King" quality while the more expensive one is "Zeus" (best)

not that it says a whole lot but ...
 

ANIA_fanboy

Distinguished
Jul 8, 2006
78
0
18,630
Yes, and the lower voltage stuff is higher quality. Increased voltage is only needed when the rated speed is an overclocked state.

I think I have to disagree with you.

Higher tested voltage allows for more overclock room.

You can run memeory at lower latency with higher voltage

Sometimes, in order to keep memory stable at lower than rated latencies you up the voltage.

Check out the additinal information on the links that I gave. It seems that they did more thourough testing on the memory with higher tested voltage to prove its potential.


On the other hand Mondoman throws a wrench in the whole discussion with his find 8O
 

rwaritsdario

Distinguished
Aug 2, 2006
3,017
0
20,780
$200 = 1.9~2.0v
$250 = 2.0~2.1v

According to crashman (and I agree with him), the cheapest is the highest quality. Something you dont see everyday.
Anyways if the more expensive one is better I dont think it would be worth $50 more since such low voltage difference shouldnt proove such high overclocking.
 

Crashman

Polypheme
Former Staff
Higher tested voltage allows for more overclock room.

No it doesn't, it means the opposite. You raise the voltage to make your overclock stable. The "test voltage" you refer to is the minimum amount the company requires to guarantee stability.

Look at it from the opposite direction: Suppose you had RAM that could run 1200 data rate at 2.20V and 800 data rate at 1.80V. Now, suppose you had RAM that REQUIRED 2.20V to get to 800 data rate in the first place...very little room to go from there.

The only major exception is RAM that overheats under higher voltage, but the this manufacturer isn't giving you the "highest voltage tolerable" it's giving you the "least voltage necessary".
 

ANIA_fanboy

Distinguished
Jul 8, 2006
78
0
18,630
Higher tested voltage allows for more overclock room.

No it doesn't, it means the opposite. You raise the voltage to make your overclock stable. The "test voltage" you refer to is the minimum amount the company requires to guarantee stability.

Look at it from the opposite direction: Suppose you had RAM that could run 1200 data rate at 2.20V and 800 data rate at 1.80V. Now, suppose you had RAM that REQUIRED 2.20V to get to 800 data rate in the first place...very little room to go from there.

The only major exception is RAM that overheats under higher voltage, but the this manufacturer isn't giving you the "highest voltage tolerable" it's giving you the "least voltage necessary".

So what is the bragging on the website about?
 

Crashman

Polypheme
Former Staff
Bragging to fanboys who don't understand the technology...

There are some new Micron chips that run cooler at high voltage and overclock well (both), but rule of thumb is for the same part, lower voltage at one speed usually means lower voltage at all speeds...so that you reach the voltage limit (imposed heat) at a higher clock speed.

The problem is, without specifying which part it is, you can only guess.
 

ANIA_fanboy

Distinguished
Jul 8, 2006
78
0
18,630
I misunderstand (being a fanboy and all).

I assumed that test voltage meant tested voltage where there is an obvious difference.

I didn't think that the test voltage was the SPD setting

Thnx for the clarification
 

Mondoman

Splendid
....

I didn't think that the test voltage was the SPD setting
...
It's not. The settings programmed into the SPD are often slower than the "max certified" settings advertised for the module. That's why you often have to set the memory timings by hand in the BIOS to get the full advertised speed.
 

Crashman

Polypheme
Former Staff
Neither. Tested voltage they specify is the voltage required to get the memory to overclock to it's rated speed. The memory comes SPD programed to run at a slower speed. That's right, this is slower memory with a "guaranteed stable" overclock speed.

People used to sell processors like that. They'd buy a bunch of Intel Celerons (300 and 333MHz) and test them to see if they could be overclocked to 450 and 500MHz at 100MHz FSB, then they'd sell them as "500MHz stable, pre-tested Celeron 333" and so forth.

This is what you get when you buy 2.2V memory, memory that you have to overclock yourself.

There's also a new tech, Enhanced Performance Profiles, that lets certain boards automatically-overclock certain RAM (if both support EPP). But it's still overclocking.

So you buy "2.2V DDR2-800", it shows up, and SPD values say it's DDR2-533, 1.80V. That's normal. You have to overclock the stuff to reach 800MHz data rate, by first raising the voltage.

Mondoman is right, I just thought I'd add detail.
 

EvilLost

Distinguished
Sep 1, 2006
31
0
18,530
Forgive me for being a bit new, but doesn't that sound like you are getting cheaper ram (at a not so cheap price?)

Or is it that they just make all the chips identically, and those that stable at higher clocks sell for more (the "premium" chips) and those that don't sell at the lower speeds ("basic" chips)?

I know they will be stable at 800mhz, (again this could be my lack of knowledge..) but do other companies do this? If I'm to buy Corsair 800mhz are they selling me chips that were made at a guaranteed 667 and are stable at 800 so they sell them at 800?

hope all that made sense
 

Crashman

Polypheme
Former Staff
Everyone is "doing it", selling DDR2-533 and DDR2-667 as DDR2-800 and up. They program it to boot at low speed because it needs higher-than-default voltage to be overclocked to the rated speed.

Even Corsair does this.