I was wondering if anyone out there could help de-mystify the difference between video memory and the system memory.
I'm working with a photo-editing app and I need more power but I'm not sure whether to spend the money on a video card or more ram. For instance, when I'm editing a photo, obviously the app reads the disk file (say, a .jpg) into memory. Now, it's my understanding that the photo is read into Ram, but the OS (xp in this case) feeds the monitor display with the video mem (I'll call it Vram). So, when I'm actually editing the photo, are my changes being physically edited in the vram, or just the ram?
My assumption is that the editing takes place in ram, and the vram is there just to then display what's on the screen. But then it seems like there's redundant stuff going on--every change in ram then must be transferred to vram since I see the edits on the screen. Can anyone give me a technical explanation of what goes on here with regard to how the pixels are moved around?
My practical issue is that when I, say, copy/paste a section of a photo, I'm experiencing too much hourglass time. I'll do a selection, hit ctrl-c, then I get the hourglass for 8-10 seconds, then I paste--and another delay. The pictures are not huge compared to what pros might work with--these are .jpegs and they're about 15 meg, but I guess in reality they're roughly 72 Meg (they're about 4000x6000 x 24bit)--but my display is 1152x864 so the whole thing never is showing at once. Would a better video card help, or more ram? I have alot of ram--512meg, and my video card is an ati 9000 with, I think 64 meg. Could the video chip be the bottlneck here--or is the system cpu what's actually moving the pixels around? I really appreciate any insight into this...
Thanks,
--jsteph
I'm working with a photo-editing app and I need more power but I'm not sure whether to spend the money on a video card or more ram. For instance, when I'm editing a photo, obviously the app reads the disk file (say, a .jpg) into memory. Now, it's my understanding that the photo is read into Ram, but the OS (xp in this case) feeds the monitor display with the video mem (I'll call it Vram). So, when I'm actually editing the photo, are my changes being physically edited in the vram, or just the ram?
My assumption is that the editing takes place in ram, and the vram is there just to then display what's on the screen. But then it seems like there's redundant stuff going on--every change in ram then must be transferred to vram since I see the edits on the screen. Can anyone give me a technical explanation of what goes on here with regard to how the pixels are moved around?
My practical issue is that when I, say, copy/paste a section of a photo, I'm experiencing too much hourglass time. I'll do a selection, hit ctrl-c, then I get the hourglass for 8-10 seconds, then I paste--and another delay. The pictures are not huge compared to what pros might work with--these are .jpegs and they're about 15 meg, but I guess in reality they're roughly 72 Meg (they're about 4000x6000 x 24bit)--but my display is 1152x864 so the whole thing never is showing at once. Would a better video card help, or more ram? I have alot of ram--512meg, and my video card is an ati 9000 with, I think 64 meg. Could the video chip be the bottlneck here--or is the system cpu what's actually moving the pixels around? I really appreciate any insight into this...
Thanks,
--jsteph