Oreo[SS]
04-27-2003, 09:56 PM
Best Bang for the Buck – A Video Card Comparison (http://www.shamikaserver.com/modules.php?op=modload&name=Diner_Wrapper&file=index&req=ShowFile&file_wrap=html/ssreviews/videocomp.html)
Let me start by saying that this comparison began during the fall of 2002 (Late August) and sat quietly in hibernation until a completely hacked version was released earlier this week. It is plain to see that the editor/rewriter named Various didn’t understand the concept with which this comparison was written. The mind set behind the comparison was to allow budget minded consumers and gamers the opportunity to see how their current video card solution stacked up against the newer cards available at the time. Please note that I said “AT THE TIME”. Yes there are newer drivers and newer cards available and I will address those issues later here in my post. The ATI Radeon 9700pro for example was provided by ATI, whom I would like to thank, shortly after its public release. Due to their other responsibilities to other review sites and the newness of the card it had to be returned so it could be passed along. Hence the difference of drivers used for the 9500pro during benchmarks. I used the best available driver at the time of the testing. Changing drivers with every new release would only provide a clouded picture of each cards ability. By running the same driver I effectively leveled the playing field across the board. Please bare this in mind. Also other project, reviews and responsibilities to my family and work added to the length of time it took to complete the review. Each of the manufacturer-supplied cards was tested for 7-10 days for performance and stability under real world conditions. This too lengthened the time of completion.
You will have noticed that the complete version of the comparison includes several sections that were conveniently left out of the hacked revision. These sections include information about issues with the ATI 9700pro and 9500pro, Visual quality, Manufacturer background and direct linkage to the manufacturers site for the product being compared and our overall opinion of each of the cards during the test period. I personally feel that these are important areas that help to give you as a reader a better understanding of each cards performance. Benchmarks are great and big numbers are awesome but if a card has issues why spend your hard earned money for something that will cause you headaches and down time. This comparison isn’t a “New Front Page Release” item for ShamikaServer.com but is more suited as a solid addition for our review/article archives. We understand that you as a reader want up-to-date information running current drivers and platforms however the comparison does allow for an overall view of the types of performance gained by upgrading from an older outdated video card solution.
Now to handle several of the questions posed in several forums around the web. Top of the list….
Q: Why wasn’t Oreo listed as writer?
Simply in comparing the two articles, their version no longer represented my vision and is now a product of Various.
Q: Why are your Parhelia numbers are out of whack?
Well they really aren’t out of whack. If you are talking about the poor performance in generating frame rates this is consistent not only in our testing rig but also in Windsorfox’s new system powered by a 2800+ Batron. The Parhelia isn’t a speed king by any means but it does offer several things you can’t get from either the TI4600’s or the Radeon 9700/9500’s such as Triple monitor display, Stereo Gaming and Visual Quality which is simply amazing. I personally own a TI4600 and Radeon 9700 but neither can touch the image eye candy produced by the Parhelia and it’s 16xAA. Am I saying the image quality of the Radeons and TI4600 is poor? Not at all, these cards offer excellent image quality to a point. Color reproduction is awesome across the board but where the Parhelia gets its edge is with the 16xAA and ability to generate fine detail in images. Of course I’m a pretty big gamer so I went for frame rate performance for my personal systems.
Q: Why not redo all the benchmarks with current drivers?
Well it would be nice however redoing the complete gamut of benchmarks will show across the board increases as both nVidia and ATI have made a lot of advancements and performance gains with the current set of drivers. Each manufacturer tweaks and pushes for that bit of an edge with the release of a new driver. Each of the tested cards is currently running in daily use systems or has been returned to the entity, which had made them available. Also just redoing lets say the Radeon benchmarks will produce a lopsided comparison. Again I was looking to keep the playing field level and not trying to prove which card was the biggest/baddest card available at the time. The use of the word SHOOTOUT implies gunning for a single winner and this isn’t a shootout it is a comparison for overall performance gains between available cards new and old for upgrading purposes.
Q: Doesn’t the Leadtek card have problems with the fans burning out on the heatsink?
Well, not to my knowledge but let me make a simple statement on this possible issue. I personally own the Leadtek WinFast A250 Ultra TD - MyVIVO Edition and it has been a rock of stability. This card resides in my main workstation and runs 24/7. If you are not familiar with this card Leadtek has placed two small dust filters to protect the heatsink and fans from the usual attack of the dust bunnies. This is a great option however as with any filter (car, lawnmower, AC unit) you have to maintain the filters and this means cleaning them. It’s quick and easy and if done regularly I foresee no issues. I clean my systems out about once every two months and it is amazing the amount of air and dust a good case moves over its components.
Q: Why are the 3Dmark Scores so low?
Most sites run top of the line state of the art systems for their reviews. In most of my reviews I try to show performance by running solid middle of the road systems. What good is having a video card comparison that runs a 3.0 GHz Intel system for people looking to upgrade a card in their older setup? Yes this can also be said for running the comparison on the dual setup however, of all the benchmarks run only Quake III and DroneZ utilize the full ability of the dual processor setup. The point of the comparison wasn’t to generate huge numbers; the point was to show you the different levels of performance between cards. Another factor for the lower score (and we are not talking a huge difference here) is the MPX chipset used on the dual AMD platform. The MPX chipset is easily outpaced by the KT300 or newer chipset platform.
Q: How can Geforce2 out perform a 9700 Pro in some tests?
Unless I missed it the Gainward Geforce2 Ti450 didn’t out perform the Radeon 9700pro in any of the benchmarks. However it did outpace the 9500pro in several and only at the lower resolutions. The ATI products do not handle OpenGL as well as nVidia products. This is common knowledge. I attribute the newness (first release) of the 9500pro’s drivers in part to the fact that it was outpaced in several benchmarks. Take a close look at the Quake III results. Again running in an OpenGL environment at 640x480 these results are really less about the video card and more about CPU/System performance at this resolution. The simplicity of the Quake engine and the ability for just about any current video cards to generate huge frame rates is one of the reasons I rarely include these results in a video card review. Don’t get me wrong the Quake III benchmark has filled its place in history but using it as a true indicator of performance with the current line of video cards available today would be much like comparing apples to oranges.
Q: Do all the Radeon 9700/9500’s have rendering issues?
This is one of those things that concerned me the most about the ATI cards. You have seen several of the images of the rendering issues these cards were having with Ghost Recon and SOFII. These at the time were MAJOR issues for me and I contacted ATI about these problems. I have to say that the engineers at ATI were not only interested but also concerned about the problems I was encountering. We talked back and forth for approximately a week working on a resolution to the issue. I was concerned that the Radeon’s may be having issues running on the dual AMD platform. ATI admitted that testing in this type of environment had been limited. In an effort to better understand if the issues are caused by the dual platform, drivers or by the card itself, I commandeered a newly built custom gaming system and installed the card. The issues remained while running in the single processor environment. At this point I was told that they (ATI) were aware of several issues with the drivers and were in the process of resolving them. It was refreshing to see that they were already working on solutions. I can also tell you as of this writing that newer revisions of the drivers have helped not only to improve performance but also helped to resolve/lessen issues encountered in games including the ones shown in the comparison.
Q: Why are your listed prices so high?
Remember the time line of this comparison. These prices were taken from pricewatch.com back in early January 2003. Another quick look at pricewatch.com really puts a smile on a persons face as much like interest rates the prices of video cards have dropped a good bit. It is a budget buyers market.
So that’s about it. If you have any more questions or comments please feel free to ask away. Even though this is just an archive review for ShamikaServer.com, I am still interested in what you would consider the Best Bang for the Buck so please vote in the poll and post your reasons why you would purchase that card.
Let me start by saying that this comparison began during the fall of 2002 (Late August) and sat quietly in hibernation until a completely hacked version was released earlier this week. It is plain to see that the editor/rewriter named Various didn’t understand the concept with which this comparison was written. The mind set behind the comparison was to allow budget minded consumers and gamers the opportunity to see how their current video card solution stacked up against the newer cards available at the time. Please note that I said “AT THE TIME”. Yes there are newer drivers and newer cards available and I will address those issues later here in my post. The ATI Radeon 9700pro for example was provided by ATI, whom I would like to thank, shortly after its public release. Due to their other responsibilities to other review sites and the newness of the card it had to be returned so it could be passed along. Hence the difference of drivers used for the 9500pro during benchmarks. I used the best available driver at the time of the testing. Changing drivers with every new release would only provide a clouded picture of each cards ability. By running the same driver I effectively leveled the playing field across the board. Please bare this in mind. Also other project, reviews and responsibilities to my family and work added to the length of time it took to complete the review. Each of the manufacturer-supplied cards was tested for 7-10 days for performance and stability under real world conditions. This too lengthened the time of completion.
You will have noticed that the complete version of the comparison includes several sections that were conveniently left out of the hacked revision. These sections include information about issues with the ATI 9700pro and 9500pro, Visual quality, Manufacturer background and direct linkage to the manufacturers site for the product being compared and our overall opinion of each of the cards during the test period. I personally feel that these are important areas that help to give you as a reader a better understanding of each cards performance. Benchmarks are great and big numbers are awesome but if a card has issues why spend your hard earned money for something that will cause you headaches and down time. This comparison isn’t a “New Front Page Release” item for ShamikaServer.com but is more suited as a solid addition for our review/article archives. We understand that you as a reader want up-to-date information running current drivers and platforms however the comparison does allow for an overall view of the types of performance gained by upgrading from an older outdated video card solution.
Now to handle several of the questions posed in several forums around the web. Top of the list….
Q: Why wasn’t Oreo listed as writer?
Simply in comparing the two articles, their version no longer represented my vision and is now a product of Various.
Q: Why are your Parhelia numbers are out of whack?
Well they really aren’t out of whack. If you are talking about the poor performance in generating frame rates this is consistent not only in our testing rig but also in Windsorfox’s new system powered by a 2800+ Batron. The Parhelia isn’t a speed king by any means but it does offer several things you can’t get from either the TI4600’s or the Radeon 9700/9500’s such as Triple monitor display, Stereo Gaming and Visual Quality which is simply amazing. I personally own a TI4600 and Radeon 9700 but neither can touch the image eye candy produced by the Parhelia and it’s 16xAA. Am I saying the image quality of the Radeons and TI4600 is poor? Not at all, these cards offer excellent image quality to a point. Color reproduction is awesome across the board but where the Parhelia gets its edge is with the 16xAA and ability to generate fine detail in images. Of course I’m a pretty big gamer so I went for frame rate performance for my personal systems.
Q: Why not redo all the benchmarks with current drivers?
Well it would be nice however redoing the complete gamut of benchmarks will show across the board increases as both nVidia and ATI have made a lot of advancements and performance gains with the current set of drivers. Each manufacturer tweaks and pushes for that bit of an edge with the release of a new driver. Each of the tested cards is currently running in daily use systems or has been returned to the entity, which had made them available. Also just redoing lets say the Radeon benchmarks will produce a lopsided comparison. Again I was looking to keep the playing field level and not trying to prove which card was the biggest/baddest card available at the time. The use of the word SHOOTOUT implies gunning for a single winner and this isn’t a shootout it is a comparison for overall performance gains between available cards new and old for upgrading purposes.
Q: Doesn’t the Leadtek card have problems with the fans burning out on the heatsink?
Well, not to my knowledge but let me make a simple statement on this possible issue. I personally own the Leadtek WinFast A250 Ultra TD - MyVIVO Edition and it has been a rock of stability. This card resides in my main workstation and runs 24/7. If you are not familiar with this card Leadtek has placed two small dust filters to protect the heatsink and fans from the usual attack of the dust bunnies. This is a great option however as with any filter (car, lawnmower, AC unit) you have to maintain the filters and this means cleaning them. It’s quick and easy and if done regularly I foresee no issues. I clean my systems out about once every two months and it is amazing the amount of air and dust a good case moves over its components.
Q: Why are the 3Dmark Scores so low?
Most sites run top of the line state of the art systems for their reviews. In most of my reviews I try to show performance by running solid middle of the road systems. What good is having a video card comparison that runs a 3.0 GHz Intel system for people looking to upgrade a card in their older setup? Yes this can also be said for running the comparison on the dual setup however, of all the benchmarks run only Quake III and DroneZ utilize the full ability of the dual processor setup. The point of the comparison wasn’t to generate huge numbers; the point was to show you the different levels of performance between cards. Another factor for the lower score (and we are not talking a huge difference here) is the MPX chipset used on the dual AMD platform. The MPX chipset is easily outpaced by the KT300 or newer chipset platform.
Q: How can Geforce2 out perform a 9700 Pro in some tests?
Unless I missed it the Gainward Geforce2 Ti450 didn’t out perform the Radeon 9700pro in any of the benchmarks. However it did outpace the 9500pro in several and only at the lower resolutions. The ATI products do not handle OpenGL as well as nVidia products. This is common knowledge. I attribute the newness (first release) of the 9500pro’s drivers in part to the fact that it was outpaced in several benchmarks. Take a close look at the Quake III results. Again running in an OpenGL environment at 640x480 these results are really less about the video card and more about CPU/System performance at this resolution. The simplicity of the Quake engine and the ability for just about any current video cards to generate huge frame rates is one of the reasons I rarely include these results in a video card review. Don’t get me wrong the Quake III benchmark has filled its place in history but using it as a true indicator of performance with the current line of video cards available today would be much like comparing apples to oranges.
Q: Do all the Radeon 9700/9500’s have rendering issues?
This is one of those things that concerned me the most about the ATI cards. You have seen several of the images of the rendering issues these cards were having with Ghost Recon and SOFII. These at the time were MAJOR issues for me and I contacted ATI about these problems. I have to say that the engineers at ATI were not only interested but also concerned about the problems I was encountering. We talked back and forth for approximately a week working on a resolution to the issue. I was concerned that the Radeon’s may be having issues running on the dual AMD platform. ATI admitted that testing in this type of environment had been limited. In an effort to better understand if the issues are caused by the dual platform, drivers or by the card itself, I commandeered a newly built custom gaming system and installed the card. The issues remained while running in the single processor environment. At this point I was told that they (ATI) were aware of several issues with the drivers and were in the process of resolving them. It was refreshing to see that they were already working on solutions. I can also tell you as of this writing that newer revisions of the drivers have helped not only to improve performance but also helped to resolve/lessen issues encountered in games including the ones shown in the comparison.
Q: Why are your listed prices so high?
Remember the time line of this comparison. These prices were taken from pricewatch.com back in early January 2003. Another quick look at pricewatch.com really puts a smile on a persons face as much like interest rates the prices of video cards have dropped a good bit. It is a budget buyers market.
So that’s about it. If you have any more questions or comments please feel free to ask away. Even though this is just an archive review for ShamikaServer.com, I am still interested in what you would consider the Best Bang for the Buck so please vote in the poll and post your reasons why you would purchase that card.