Does A High End Graphics Card Affect An Image's Render Quality?
Hello everyone.
I am trying to put together a PC that I can used for creating 3D content such as 3D images rendered from 3D Studio Max or Maya. I am somewhat confused at this point on what video card to get. I've heard from several persons that the Video card does not affect the render time of an image. Basically I've heard that the render process is all performed by the PC's CPU and that the video card only affects what you see on screen while you are designing but not the final render.
So if an image takes 16 hours to render the only way to reduce the render time is to get a faster CPU or more CPUs but not a better video card. Is this true?
Does this also mean then that the video card on the pc that is rendering an image does not affect the quality of the image being rendered? In other words, the 3d image quality of an image designed and rendered on a PC with a cheap video card such as an NVidia 8800GT(which costs about ($200) will look just as good as a pc with an expensive video card such as an NVidia Quadro FX 4800 (which costs about $1,600)? The textures such as metals and such will look just as high quality on both PCs final rendered image file (i.e. .png file formats)?
The most helpful responses to these questions for me will be those that can provide internet links to articles from reputable sources. Thanks in advance.
I am trying to put together a PC that I can used for creating 3D content such as 3D images rendered from 3D Studio Max or Maya. I am somewhat confused at this point on what video card to get. I've heard from several persons that the Video card does not affect the render time of an image. Basically I've heard that the render process is all performed by the PC's CPU and that the video card only affects what you see on screen while you are designing but not the final render.
So if an image takes 16 hours to render the only way to reduce the render time is to get a faster CPU or more CPUs but not a better video card. Is this true?
Does this also mean then that the video card on the pc that is rendering an image does not affect the quality of the image being rendered? In other words, the 3d image quality of an image designed and rendered on a PC with a cheap video card such as an NVidia 8800GT(which costs about ($200) will look just as good as a pc with an expensive video card such as an NVidia Quadro FX 4800 (which costs about $1,600)? The textures such as metals and such will look just as high quality on both PCs final rendered image file (i.e. .png file formats)?
The most helpful responses to these questions for me will be those that can provide internet links to articles from reputable sources. Thanks in advance.
ASKER
Hello pjvog.
My confusion on image quality being impacted by a video card comes from statements such as the following from the Nvidia website which I don't if is hype or not:
"128-Bit Precision Graphics Pipeline
Enables sophisticated mathematical computations to maintain high accuracy, resulting in unmatched visual quality. Full IEEE 32-bit floating-point precision per color component (RGBA) delivers millions of color variations with the broadest dynamic range."
....
"Essential for Microsoft® Windows Vista®
Offering an enriched 3D user interface, increased application performance, and the highest image quality, NVIDIA Quadro graphics boards and NVIDIA OpenGL ICD drivers are optimized for 32- and 64-bit architectures to enable the Windows Vista experience."
These Nvidia statements came from the following link under their "Overview" tab:
http://www.nvidia.com/object/product_quadro_fx_4800_us.html
Both comments refer to image quality so it makes me wonder about the truth of it. The above statements from Nvidia in regards to "Image Quality" do not appear to be subject based as you are hinting "like measuring art". The statements above I believe are referring to accuracies of colors and textures as well as color depths.
My confusion on image quality being impacted by a video card comes from statements such as the following from the Nvidia website which I don't if is hype or not:
"128-Bit Precision Graphics Pipeline
Enables sophisticated mathematical computations to maintain high accuracy, resulting in unmatched visual quality. Full IEEE 32-bit floating-point precision per color component (RGBA) delivers millions of color variations with the broadest dynamic range."
....
"Essential for Microsoft® Windows Vista®
Offering an enriched 3D user interface, increased application performance, and the highest image quality, NVIDIA Quadro graphics boards and NVIDIA OpenGL ICD drivers are optimized for 32- and 64-bit architectures to enable the Windows Vista experience."
These Nvidia statements came from the following link under their "Overview" tab:
http://www.nvidia.com/object/product_quadro_fx_4800_us.html
Both comments refer to image quality so it makes me wonder about the truth of it. The above statements from Nvidia in regards to "Image Quality" do not appear to be subject based as you are hinting "like measuring art". The statements above I believe are referring to accuracies of colors and textures as well as color depths.
ASKER
Here is another statement I found regarding the Nvidia Quadro FX 4800:
"Highest Color Fidelity - 10-bit per component color fidelity enables billions rather than millions of
color variations for rich, vivid image quality with the broadest dynamic range."
This statement comes from the following link:
http://www.nvidia.com/docs/IO/63157/NV_DS_QFX_4800_US_Nov08_LowRes.pdf
It's statements like these that make me wonder if the card actually affects the final render quality of an image. But again, I don't know if its just hype. I guess what would really help is if someone had the Quadro FX 4800 on a pc and another pc with an ordinary video card and both pcs were used to render a high res 3d image with high reflective metals, environment details, etc. And ideally both pcs would have identical cpus. The question would be if the rendered image file from both PC's is of identical quality. If render quality is all from cpu then the images should be of identical quality.
This is what I am trying to find out. If they are of identical quality then I don't think I will spend a lot of money on the video card.
"Highest Color Fidelity - 10-bit per component color fidelity enables billions rather than millions of
color variations for rich, vivid image quality with the broadest dynamic range."
This statement comes from the following link:
http://www.nvidia.com/docs/IO/63157/NV_DS_QFX_4800_US_Nov08_LowRes.pdf
It's statements like these that make me wonder if the card actually affects the final render quality of an image. But again, I don't know if its just hype. I guess what would really help is if someone had the Quadro FX 4800 on a pc and another pc with an ordinary video card and both pcs were used to render a high res 3d image with high reflective metals, environment details, etc. And ideally both pcs would have identical cpus. The question would be if the rendered image file from both PC's is of identical quality. If render quality is all from cpu then the images should be of identical quality.
This is what I am trying to find out. If they are of identical quality then I don't think I will spend a lot of money on the video card.
I read it like a gamer. Higher card performance is going to increase the likelihood of "accurate detail". I'm just laying out my thinking. Think of an older computer running a new game. Given the time required, it is impossible for the older pc to produce high quality images... in the amount of time it takes to look around in the game. I read quality with time factored in.
Where I read : "Essential for Microsoft® Windows Vista®
Offering an enriched 3D user interface, increased application performance, and the highest image quality, NVIDIA Quadro graphics boards and NVIDIA OpenGL ICD drivers are optimized for 32- and 64-bit architectures to enable the Windows Vista experience."
I think, "on an older computer, the graphics card isn't capable of running the graphic intensive windows desktop, but can display a pre-rendered high quality PNG OF that graphic intensive desktop".
Also " I guess what would really help is if someone had the Quadro FX 4800 on a pc and another pc with an ordinary video card and both pcs were used to render a high res 3d image with high reflective metals, environment details, etc. And ideally both pcs would have identical cpus. The question would be if the rendered image file from both PC's is of identical quality. If render quality is all from cpu then the images should be of identical quality. " Makes me think "Need identical monitors, colors are unique to everyone" (art measured, lol). Same as "Highest Color Fidelity", which just sounds like hype to me, new cards increase performace.
Either way, I think the only way to find out is to test it, if we're working off logic and theories alone, I only have a Quadro NVS 160M in my laptop, a 7600gt in my one box, and a 260gxt in another, we could probably create our own benchmark, heh.
Where I read : "Essential for Microsoft® Windows Vista®
Offering an enriched 3D user interface, increased application performance, and the highest image quality, NVIDIA Quadro graphics boards and NVIDIA OpenGL ICD drivers are optimized for 32- and 64-bit architectures to enable the Windows Vista experience."
I think, "on an older computer, the graphics card isn't capable of running the graphic intensive windows desktop, but can display a pre-rendered high quality PNG OF that graphic intensive desktop".
Also " I guess what would really help is if someone had the Quadro FX 4800 on a pc and another pc with an ordinary video card and both pcs were used to render a high res 3d image with high reflective metals, environment details, etc. And ideally both pcs would have identical cpus. The question would be if the rendered image file from both PC's is of identical quality. If render quality is all from cpu then the images should be of identical quality. " Makes me think "Need identical monitors, colors are unique to everyone" (art measured, lol). Same as "Highest Color Fidelity", which just sounds like hype to me, new cards increase performace.
Either way, I think the only way to find out is to test it, if we're working off logic and theories alone, I only have a Quadro NVS 160M in my laptop, a 7600gt in my one box, and a 260gxt in another, we could probably create our own benchmark, heh.
Hello zonkerman,
As a video games programmer I can say YES the quality of your graphics card does affect the quality of the final image displayed on screen.
Colour depth, anti-aliasing, the number of polygons that can be rendered, there are a lot of features that some games will use if they are available. A card that is Direct X 11 compliant is the latest best standard.
Remember some games will not use all the features of some cards. It depends when they were written and what specification they were written to. Some games require the latest technlogy and some don't.
All the best
Sci-Fi Si
As a video games programmer I can say YES the quality of your graphics card does affect the quality of the final image displayed on screen.
Colour depth, anti-aliasing, the number of polygons that can be rendered, there are a lot of features that some games will use if they are available. A card that is Direct X 11 compliant is the latest best standard.
Remember some games will not use all the features of some cards. It depends when they were written and what specification they were written to. Some games require the latest technlogy and some don't.
All the best
Sci-Fi Si
ASKER CERTIFIED SOLUTION
membership
This solution is only available to members.
To access this solution, you must be a member of Experts Exchange.
SOLUTION
membership
This solution is only available to members.
To access this solution, you must be a member of Experts Exchange.
ASKER
Thanks for all the input so far guys.
The information on performance was interesting however please note that the main question I was trying to get answered was about image quality. Some of you have provided intormation in this regard and I thiank you. The CPU i'm will be using is an Intel I7 950 on a pc with 12 gigs of Corsair Dominator memory. The video card so far on my target list is an NVidia Quadro 4800 for applications such as 3ds max. Gaming is not a priority, although I would like to play a little when free time permits. But first and formost of interest is producing high quality 3d images with the most realistic textures and environment elements. Whether rendering a single such image take 1 hour or 6 hours is not as much a concern to me as whether or not the image colors and textures look as real and refined as possible.
Hense, it is the quality of image produced, not the speed, that I am wondering as to whether the video card affects or not. Some have already provided their feedback to this and again I thank you for your input. I will keep this thread going a little longer to see what others have to say.
The information on performance was interesting however please note that the main question I was trying to get answered was about image quality. Some of you have provided intormation in this regard and I thiank you. The CPU i'm will be using is an Intel I7 950 on a pc with 12 gigs of Corsair Dominator memory. The video card so far on my target list is an NVidia Quadro 4800 for applications such as 3ds max. Gaming is not a priority, although I would like to play a little when free time permits. But first and formost of interest is producing high quality 3d images with the most realistic textures and environment elements. Whether rendering a single such image take 1 hour or 6 hours is not as much a concern to me as whether or not the image colors and textures look as real and refined as possible.
Hense, it is the quality of image produced, not the speed, that I am wondering as to whether the video card affects or not. Some have already provided their feedback to this and again I thank you for your input. I will keep this thread going a little longer to see what others have to say.
Still Image quality may well have more to do with your monitor than your graphics card. I don't think there is a lot of variation of stills produced by the graphics card but at the monitor end the levels of quality are hugely different.
Zonkerman -
There is a British magazine called 3D World, it's not a common newstand title in the US (I'm not sure where you are located), but they put most of their content on their website 3dworldmag.com.
The OCtober 2009 issue has a pretty good article on customizing workstations, that goes into some explanation of how your machine would differ if you were primarily rendering, vs primarily modelling, or primarily compositing. Its worth a read if you can find the article. Even if the specific specifications they suggest are not what you end up with, it's an interesting insight into how the different functions within the whole visualization workflow utilize different aspects of a computer.
There is a British magazine called 3D World, it's not a common newstand title in the US (I'm not sure where you are located), but they put most of their content on their website 3dworldmag.com.
The OCtober 2009 issue has a pretty good article on customizing workstations, that goes into some explanation of how your machine would differ if you were primarily rendering, vs primarily modelling, or primarily compositing. Its worth a read if you can find the article. Even if the specific specifications they suggest are not what you end up with, it's an interesting insight into how the different functions within the whole visualization workflow utilize different aspects of a computer.
That's one heck of a machine, what kind of monitor will you be using?
You have exactly the same spec as I have machine -wise and i can guarantee your graphics card is not going to be an issue. Take a look at your monitor/s :)
ASKER
"What kind of monitor will you be using?"
Well, I actually have not yet picked a new one to purchase. I was going to try and use my existing monitor on my old system to see if it works well enough. I had bought it a few months ago so I would like to at try it on the new system. I did not pay much for it, about $300, but again it was for my old system so did not want to invest much. That was before I decided to just get another system. My existing monitor is as follows:
Manufacturer: LG
Model: Flatron W2453V
24 inch widescreen (16:9)
Pixel Pitch: 0.277mm
Display Colors: 16.7 Million
Brightness : 300 cd/m2
Input Video Compatibility: Analog RGB, Digital (I'm using a DVI Cable)
Response: 2ms
Digital Contrast Ratio: 50,000:1
Resolution: 1920x1080P (Full HD)
Connectors : D-Sub, DVI-D
D-Sub : 1
DVI : 1
HDMI : 1
I trying to keep costs down on my new system, below or around $4,000, so the monitor is pretty much out of my initial purchase scope.
Well, I actually have not yet picked a new one to purchase. I was going to try and use my existing monitor on my old system to see if it works well enough. I had bought it a few months ago so I would like to at try it on the new system. I did not pay much for it, about $300, but again it was for my old system so did not want to invest much. That was before I decided to just get another system. My existing monitor is as follows:
Manufacturer: LG
Model: Flatron W2453V
24 inch widescreen (16:9)
Pixel Pitch: 0.277mm
Display Colors: 16.7 Million
Brightness : 300 cd/m2
Input Video Compatibility: Analog RGB, Digital (I'm using a DVI Cable)
Response: 2ms
Digital Contrast Ratio: 50,000:1
Resolution: 1920x1080P (Full HD)
Connectors : D-Sub, DVI-D
D-Sub : 1
DVI : 1
HDMI : 1
I trying to keep costs down on my new system, below or around $4,000, so the monitor is pretty much out of my initial purchase scope.
I use a Flatron w23 as my secondary monitor. I definitely like, although I find the 16:9 restricting after my using my primary for so long (a 16:10, res @ 1920x1200 instead of 1920x1080)
My Primary is a Samsung T260 (25.5") which I love, very much.
I'm sorry I can't help as much as others as far as your original question, but I at least experiment with monitors a lot, lol.
My Primary is a Samsung T260 (25.5") which I love, very much.
I'm sorry I can't help as much as others as far as your original question, but I at least experiment with monitors a lot, lol.
99.9% of the time when you render out an image from maya you should use maya software render as opposed to maya hardware. The maya hardware option is reliant on shaders in the card. This is fine for games development as the games are dependent on the same shaders. When it comes to film production the visual effects probably go beyond what the hardware shaders can accomplish so they only real option is software rendering. This is entirely dependant on processing power. some specs for homemade renderfarms dont even have a high end 3d card to keep costs down.
see
http://helmer.sfe.se/
for some ideas
see
http://helmer.sfe.se/
for some ideas
Maya support many renderers. So some of them (maya hardware render, for example) use openGL.
Maya Software Render - doesn't use.
Mental Ray renderer can use Harware shaders, but usually not...
Maya Software Render - doesn't use.
Mental Ray renderer can use Harware shaders, but usually not...
I doubt that final quality is determined by the price of the video card, I think the point of getting a better card is to improve the speed at which you render the image. Googling a few choice terms brings up quite a few results using phrases like "High end graphics provide near real-time rendering".
From the other EE post, the link to Maya's qualified hardware list : http://usa.autodesk.com/adsk/servlet/pc/item?id=13583898&siteID=123112#QUAL/maya_80_win.html
The info in that link says that Maya will only use what's on that list to Optimally run. From a personal standpoint of computer gaming, I think about the purpose of having professional workstation cards.
Oh, and all the comparison charts I ever see, compare the SPEED/Performance of a graphics card when rendering : http://www.nvidia.com/object/IO_11761.html. I don't know if there's a way to measure image quality, that's like measuring art.