Sep152012

Is compromising on dedicated video memory right while upgrading to a higher series of a graphics card?

Osama Javaid asks:

Is it a good decision to upgrade from a 4GB DDR3 GT series card to a 1 GB GDDR5 GTX series card ? I mean are there any cons?


Browser: Firefox 15.0.1
System: Windows 7 Ultimate 64-bit SP1
Tagged: , , , ,

Comments for this Question are closed.

If you are looking for help, please ask a new question here.

We will be happy to help you!

4 Answers -

0 votes

Mike

September 15, 2012

In general anything between 1-2GB is fine. The only case where you’d benefit from more dedicated memory is probably some GPU based 3D rendering (3DSM, Blender) or something like that.

However, you will have to be more specific about the models if you want a better answer. Without knowing the cards and making an actual comparison I wouldn’t recommend a switch from GT to GTX.

0 votes

ha14

September 15, 2012

GB GDDR5 GTX should be better than 4GB DDR3 GT, sometime can be little difference, but then testing will be needed.

GT 650M GDDR5 vs DDR3 and 2GB vs 1GB memory capacity
http://www.gaminglaptopsjunky.com/gt-650m-gddr5-vs-ddr3-impact-gt-650m-2gb-1gb-memory-capacity/

0 votes

Paul Girardin

September 16, 2012

From here, it does not look like a compromise!

Of course, your mileage may vary depending on the card’s specific details (Vendor, quality, ETC …)

GDDR5 can for all purpose be considered at least twice as fast than GDDR3.

GDDR3 Bandwidth: up to 9.6 Gbytes/s with a 1GHz clock.

GDDR5 operates at up to 5Gbps at 1.5 volts.

GDDR5 operates with two different clock types. A differential command clock (CK) to where address and command inputs are referenced, and a forwarded differential write clock (WCK) where read and write data are referenced to.

Being more precise, the GDDR5 SGRAM uses two write clocks, each of them assigned to two bytes.

The WCK runs at twice the CK frequency.

Taking a GDDR5 with 5 Gbps data rate per pin as an example, the CK clock runs with 1.25 GHz and WCK with 2.5 GHz.

The CK and WCK clocks will be aligned during the initialization and training sequence.

This alignment allows read and write access with minimum latency.

See here for details: http://www.interfacebus.com/GDDR5-Memory-IC-Pinout.html

And here to start exploring further on the subject: http://en.wikipedia.org/wiki/GDDR5

Hope this helps

Osama Javaid

Too much over-techyy answer, can’t understand

September 16, 2012
Paul Girardin

Simple answer: Less dedicated video memory on a better GDDR5 type card is not a compromise, it’s an improvement!

You need less memory to do the job twice as fast than GDDR3!

September 16, 2012
0 votes

Dalsan

September 18, 2012

The obvious con would be lower memory, not that it would matter that much unless you actually use programs that can utilize more than 2GB, but you would gain better compatibility and performance with newer games and other software that are GPU capable. The comparison would be like having a 2.0GHz dual core Pentium (Core Duo) with 8GB of ram versus a 2.0GHz Core i3 with 4GB ram. In this case, most times the lesser amount of ram doesn’t equate to worse performance unless you use programs that can utilize the extra ram, but still, performance drops are less noticeable than if it were the same processors with less ram than the other. In other words, more horsepower and torque with less storage capacity will most times beat less horsepower and torque with larger storage capacity. Only suggestion would be to try to get a GT model over the GTX model.