Find solution
Does Zoloft affect memory?

How does the Memory on a GPU affect performance?

  • For example when a GPU has 1GB or 2GB of GDDR5 Memory, what purpose does this serve, does it affect FPS? Or graphics? And on the side can you please explain to me the purpose of Memory Clock Speed , usually around 4000-6000 mhz. Thanks

  • Answer:

    Having additional video memory on a graphics card allows the card to run at higher resolutions without suffering a dramatic performance dropoff. It also helps if you want to crank up the detail level, since video memory is used to store textures. However, there's a disconnect between public perception and what's really useful- and card manufacturers use these widespread misconceptions to their advantage. For example, the average consumer who doesn't know much about graphics cards has probably heard many times "buy a 1gb card", sometimes even by so-called experts. But the TYPE of memory (i.e. fast GDDR5 vs slower GDDR3 or GDDR2) has far greater impact than the amount. People assume 1gb cards are much better than 512mb models, but that's not true. The amount of video memory on a graphics card has never been a direct measure of performance! It's the GPU a card is based upon that determines performance, not the RAM. A 512mb Radeon HD 5670 is still 4X faster than a 1gb (or even 2gb) Radeon HD 5450. http://www.techspot.com/review/245-ati-radeon-hd-5570/page4.html At resolutions of 1680x1050 or lower, 512mb and 1gb versions of same card perform exactly the same. At 1920x1080 having 1gb helps keep gameplay smooth. As a result, putting 1gb on any card below a Radeon HD 6770 or GeForce GTX 550 Ti (the minimum cards needed to play most games at full 1080p) is useless. For resolutions above 1920x1080 (like 2560x1600 or when running multiple monitors) that's when having a 2gb card helps. If graphics cards were automobiles, the GPU would be engine while the RAM is the size of the gas tank. Having a bigger gas tank allows a car to go farther without having to stop and refill, but it doesn't make the card faster on the highway- having a V8 instead of a V4 does that. By the same token, more video RAM lets a card go farther (to higher resolutions) but doesn't fundamentally make it faster. A Corvette with a 25-gallon tank beats a station wagon with a 40-gallon tank from L.A. to Houston because the sportscar hits much higher speeds on the highway, even having to stop and refuel. And for shorter trips (i.e. lower resolutions) having the larger gas tank doesn't help at all. Memory clock speeds and core speeds do affect performance- it's the manufacturers tweaking things for optimal performance (like the air/fuel mixture). For example an overclocked card may be 10% faster than a stock model... but it's not as big of a difference as moving up to a higher-end card with a more powerful GPU. An overclocked GTS 450 is still slower than a stock GTX 550 Ti. An overclocked GTX 550 Ti still doesn't match a stock GTX 460. Here's look at how 512mb and 1gb versions of the Radeon HD 4870 cards compared to to the GeForce GTX 260.. as you can see, the differences due to memory were very small- in fact the difference isn't due to the memory size disparity at all, but simply that 1gb versions of the cards were clocked slightly higher than 512mb versions. http://www.xbitlabs.com/articles/graphics/display/radeon-hd4870-1024_14.html#sect0 So in a nutshell- more than 1gb of video memory is helpful at really high resolutions. Some newer game titles use tessellation, which means they can also benefit from having >1gb of video memory even at 1920x1080. Crysis 2 introduced a texture pack which increased the usage of video memory, allowing 1.5gb and 2gb cards to perform more smoothly than 1gb models. But the impact of more video memory is almost always less than the impact of a higher-end card model.

C-Man at Yahoo! Answers Visit the source

Was this solution helpful to you?

Other answers

Memory, GDDR5, is used for things such as storing information to display an image to the screen, buffer information, and Z coordinate information (3D games). So while it won't directly affect performance it will to a point. Clock speed is what's important, this is what helps solve the problems and send them to the pipelines where 3D graphic equations are sent to be analyzed and transformed to a 2D pixel format for display to the screen. For a little more info this wiki entry helps: http://en.wikipedia.org/wiki/Video_card

Teresa

Well the memory stores video data. The capacity affects how much your video data your card can store. More is better, BUT just as like system ram, you only need as much as you will be using. Having more then you need does not help at all. For the most part 1-1.5gb is enough to be able to play games at 1080P at max settings just fine w/o losing performance. 2gb+ is only useful when using multi displays. As far as the memory, there is bus width (eg 128bit), the memory clock speed, and the memory type (eg gddr5). The memory type works the same way as it does on ram. The higher it is, the faster the ram will be able to be. Like gddr5 can have faster memory clocks then a gddr3 is capable of. The memory clock speed is the speed at which data can go in/out of the memory. So having faster memory could help when you dont have as high of capacity, as you will be able to throw out old data and get new data faster, requiring less capacity needed. The memory bus width goes along the same lines. The bus width is how much data can go in at one time. So the clock speed is the speed at which the data is traveling, the bus width is how much data can travel at once. So having a higher bus width will be able to allow for the same bandwidth at slower clock speeds. But majority of the card performance is based on the number of core processors/cuda cores/shaders (all different names for the same thing). These are what do the rendering of the video, the more you have, the more you are able to render at one time. And there is the core and shader clock speeds aswell. The core clock is the clock speed of the GPU core itself. the GPU is what does all the mathematical calculations and float points in video rendering. The shader clock is the speed at which the processing cores/shaders run at. So a faster clock will have the same performance with less cores/shaders. (amd cards link the core clock and shader clock. so on an amd card 800mhz core clock = 800mhz shader clock. nvidia have shader clocks separate, and run them at MUCH higher rates, as nvidia cards use higher clocks less cores/shaders).

Jeff

Related Q & A:

Just Added Q & A:

Find solution

For every problem there is a solution! Proved by Solucija.

  • Got an issue and looking for advice?

  • Ask Solucija to search every corner of the Web for help.

  • Get workable solutions and helpful tips in a moment.

Just ask Solucija about an issue you face and immediately get a list of ready solutions, answers and tips from other Internet users. We always provide the most suitable and complete answer to your question at the top, along with a few good alternatives below.