GFFX 500MHz DDR-II Latency

[Dave Baumann]

I think the 500MHz DDR-II RAM suffers quite heavily from latency issues at low resolution.

I've just run some 3DMark2001 PS tests and in the lower resolutions running @ 400/400 is faster than running @ 500/500. Interestingly, Humus's front-to-back test shows a higher pixel rejection efficiency at 400/400 than 500/500, and I can't help thinking its related.

 

[Nagorak]

 

[Ichneumon]

So you think the memory controller is changing the memory latency timings dynamically depending on the clock speed?

Seems like that would be somewhat strange... I guess it would make sense in a card that is designed to change its clock speeds regularly, but not something I would have expected.

 

[BRiT]

I'm sure you've ruled this out already, but just have to ask... Are you sure that the card isn't overheating and falling back to 2D (300Mhz) speed when you're attempting to run anything at Ultra (500Mhz) speeds?

 

[Mulciber]

Have you tried running it at 500/400?

 

[Crusher]

Have you tried feeding it pudding? Maybe it's supposed to run on pudding. I like vanilla. Use vanilla. And hang it from the ceiling first, in case sitting on the floor makes it feel short.

 

[Simon F]

So you think the memory controller is changing the memory latency timings dynamically depending on the clock speed?

It's probably the driver and it's not surprising.

The page break delays for SDRAM/DDRAM and presumably the newer variants is a fixed amount of time, not a fixed number of clock cycles. If you increase the clock rate you have to increase the number of clocks for a page break (to stay within the specifications),

 

[Dave Baumann]

Are you sure that the card isn't overheating and falling back to 2D (300Mhz) speed when you're attempting to run anything at Ultra (500Mhz) speeds?

The fan goes into silent mode when in 2D speeds, so its easy to tell!

Have you tried feeding it pudding?

It spits it back out.

 

[martrox]

Have you tried feeding it pudding?

It spits it back out.

BAD BABY! BAAD BABY!!!!!

 

[Fuz]

Damn the FX, I thought she swallowed.

 

[Joe DeFuria]

(Fears this might spawn another round of FX / Photoshop art....)

 

[martrox]

Damn the FX, I thought she swallowed.

ACK! Maybe this should be BeyondXXXD!

 

[epicstruggle]

Damn the FX, I thought she swallowed.

ACK! Maybe this should be BeyondXXXD!

The address is available on register.com.
later,

 

[Tagrineth]

Damn the FX, I thought she swallowed.

::giggles::

 

[Unit01]

LOL
Wouldn't 450mhz ddr1 be more useful than ddrII500, if what you say is true?

 

[RussSchultz]

If you could find DDR-I that runs at those, it might be.

 

[arjan de lumens]

Ddin't Hynix recently announce 500MHz DDR-I?

 

[SirPauly]

SEOUL, Korea, December 10, 2002 – Hynix Semiconductor Inc. today announced it has developed a 500MHz-128Mb (4Mx32) DDR-I SDRAM for supply to the graphic chipset manufacturing market.

This high speed DDR-I SDRAM product, an industry first, features 500MHz frequency ( 1Gbps) speed, 32bit wide I/O and 144-ball FBGA package type. The new product supports the high speed needs of desktop and work station graphics and switch and router applications.

The new product has the highest speed in the DDR-I product line and is compatible with the DDR-I board. Hynix forecasts the combination of speed, compatibility, and an estimated cost advantage of 10% over DDR-II, will gain quick acceptance for the new product from the graphics and network industries. Farhad Tabrizi, Vice President of Worlwide Marketing, stated "providing the highest performance at the lowest possible price is the key for differentiating the product by end application. DDR-I at 500 MHz does just that."

 

[PiNkY]

The fan goes into silent mode when in 2D speeds, so its easy to tell!

It would not make much sense to spin down the fan, if the card overheats and thus downclocks...

 

[Jabjabs]

PiNKY, that is the name I've been stuck with for over 10 years now (thanks to some ungrateful friends)!!!