Got a tip for us?
Let us know
Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
Video Card Options for 2008 Mac Pro
- Thread starter LiamHD
- Start date
- Sort by reaction score
You are using an out of date browser. It may not display this or other websites correctly.
You should upgrade or use an alternative browser.
You should upgrade or use an alternative browser.
I just wanted to chime in with the following advice. Don't bother to get anything newer than the GT120, it's a waste of money.
I have a 2008 Mac Pro, which I upgraded to a Radeon 5770 back in 2011. That upgrade was great, much better performance than the stock 2600. Earlier this week, I decided to upgrade to the Radeon 7950 Mac Edition because I was looking to double the performance of a game that was struggling to maintain 30fps on my machine and I was hoping for something that could hit an even 60fps. The 7950 seemed like a good choice, since benchmark scores are roughly 3x what the 5770 has.
As it turns out, I barely saw any improvement at all. The game in question continued to hover around 30fps, although the Windows version was getting closer to 40fps. This really surprised me, because I was honestly expecting a noticeable difference. For further verification that it wasn't the game, I decided to run the popular Heaven benchmark. I got a score of 426 from the Radeon 7950. This is much lower than the average of 1300+ that the 7950 is supposed to get from the benchmark. The card itself works fine, but I couldn't justify the $400 price for a card that provided no noticeable performance improvements so I ended up returning it and putting the 5770 back in.
You made a good choice with the GT120, because anything newer might have been a waste. Just wanted to share this story so someone else doesn't make the same mistake. (From what I've read, the newer models of Mac Pro have PCI Express 2.1 instead of 2.0 like the 2008 Pro, so they do benefit from the better card.)
I have a 2008 Mac Pro, which I upgraded to a Radeon 5770 back in 2011. That upgrade was great, much better performance than the stock 2600. Earlier this week, I decided to upgrade to the Radeon 7950 Mac Edition because I was looking to double the performance of a game that was struggling to maintain 30fps on my machine and I was hoping for something that could hit an even 60fps. The 7950 seemed like a good choice, since benchmark scores are roughly 3x what the 5770 has.
As it turns out, I barely saw any improvement at all. The game in question continued to hover around 30fps, although the Windows version was getting closer to 40fps. This really surprised me, because I was honestly expecting a noticeable difference. For further verification that it wasn't the game, I decided to run the popular Heaven benchmark. I got a score of 426 from the Radeon 7950. This is much lower than the average of 1300+ that the 7950 is supposed to get from the benchmark. The card itself works fine, but I couldn't justify the $400 price for a card that provided no noticeable performance improvements so I ended up returning it and putting the 5770 back in.
You made a good choice with the GT120, because anything newer might have been a waste. Just wanted to share this story so someone else doesn't make the same mistake. (From what I've read, the newer models of Mac Pro have PCI Express 2.1 instead of 2.0 like the 2008 Pro, so they do benefit from the better card.)
I just wanted to chime in with the following advice. Don't bother to get anything newer than the GT120, it's a waste of money.
I have a 2008 Mac Pro, which I upgraded to a Radeon 5770 back in 2011. That upgrade was great, much better performance than the stock 2600. Earlier this week, I decided to upgrade to the Radeon 7950 Mac Edition because I was looking to double the performance of a game that was struggling to maintain 30fps on my machine and I was hoping for something that could hit an even 60fps. The 7950 seemed like a good choice, since benchmark scores are roughly 3x what the 5770 has.
As it turns out, I barely saw any improvement at all. The game in question continued to hover around 30fps, although the Windows version was getting closer to 40fps. This really surprised me, because I was honestly expecting a noticeable difference. For further verification that it wasn't the game, I decided to run the popular Heaven benchmark. I got a score of 426 from the Radeon 7950. This is much lower than the average of 1300+ that the 7950 is supposed to get from the benchmark. The card itself works fine, but I couldn't justify the $400 price for a card that provided no noticeable performance improvements so I ended up returning it and putting the 5770 back in.
You made a good choice with the GT120, because anything newer might have been a waste. Just wanted to share this story so someone else doesn't make the same mistake. (From what I've read, the newer models of Mac Pro have PCI Express 2.1 instead of 2.0 like the 2008 Pro, so they do benefit from the better card.)
You are a victim of Apple's indifference.
You can toss a couple kexts and DOUBLE your FPS.
This is a WELL known issue at Netkas, go have a gander.
I just wanted to chime in with the following advice. Don't bother to get anything newer than the GT120, it's a waste of money.
I have a 2008 Mac Pro, which I upgraded to a Radeon 5770 back in 2011. That upgrade was great, much better performance than the stock 2600. Earlier this week, I decided to upgrade to the Radeon 7950 Mac Edition because I was looking to double the performance of a game that was struggling to maintain 30fps on my machine and I was hoping for something that could hit an even 60fps. The 7950 seemed like a good choice, since benchmark scores are roughly 3x what the 5770 has.
As it turns out, I barely saw any improvement at all. The game in question continued to hover around 30fps, although the Windows version was getting closer to 40fps. This really surprised me, because I was honestly expecting a noticeable difference. For further verification that it wasn't the game, I decided to run the popular Heaven benchmark. I got a score of 426 from the Radeon 7950. This is much lower than the average of 1300+ that the 7950 is supposed to get from the benchmark. The card itself works fine, but I couldn't justify the $400 price for a card that provided no noticeable performance improvements so I ended up returning it and putting the 5770 back in.
You made a good choice with the GT120, because anything newer might have been a waste. Just wanted to share this story so someone else doesn't make the same mistake. (From what I've read, the newer models of Mac Pro have PCI Express 2.1 instead of 2.0 like the 2008 Pro, so they do benefit from the better card.)
http://netkas.org/?p=1325
As I mentioned in my post, I had also tried on Windows and the improvement was only around 10fps. I doubt deleting the Kexts on OS X will improve performance under Windows. I did install the most recent Windows drivers after installing the card.You are a victim of Apple's indifference.
You can toss a couple kexts and DOUBLE your FPS.
This is a WELL known issue at Netkas, go have a gander.
A 30 to 40% improvement for a card 3x as powerful isn't worth it. Something else had to be happening.
As I mentioned in my post, I had also tried on Windows and the improvement was only around 10fps. I doubt deleting the Kexts on OS X will improve performance under Windows. I did install the most recent Windows drivers after installing the card.
A 30 to 40% improvement for a card 3x as powerful isn't worth it. Something else had to be happening.
I have a 3,1 and can verify this works.
In the time it took you to say you won't try it, you could have.
Don't be afraid, it makes a HUGE difference. (only with 3,1 and AMD 7xxx card)
Last edited:
I have used the HD2600XT, 8800GT, HD4870 and HD5870 cards in my 3,1.
The performance ratio seems to be 2600XT was a third of the performance of the 8800GT, the 4780 was double the 8800GT and the 5870 double the 4870.
I used all of them to play WoW over 5 years and the environment detail, distance and fps have all increased in line with the above ratios.
In more serious image work it seems to follow the same path.
The performance ratio seems to be 2600XT was a third of the performance of the 8800GT, the 4780 was double the 8800GT and the 5870 double the 4870.
I used all of them to play WoW over 5 years and the environment detail, distance and fps have all increased in line with the above ratios.
In more serious image work it seems to follow the same path.
You missed the part of my post where I said that I returned the card because of this problem. It's pretty hard to test when I don't have the card anymore.I have a 3,1 and can verify this works.
In the time it took you to say you won't try it, you could have.
Don't be afraid, it makes a HUGE difference. (only with 3,1 and AMD 7xxx card)
For what it's worth, I think the old CPU was also a contributing factor in the cases I was testing (minus the benchmark, which claims to be entirely GPU driven). One of my tests was with the Gamecube/Wii emulator Dolphin, both the Windows and Mac versions. In that case, the times it dropped below 60fps were identical to the old card and when I finally thought to check CPU use, 2 cores were maxed out. Another test I ran was with a beta of a cross-platform game designed for the Oculus rift. (I don't have one, but it was one of my motivations for wanting a faster graphics card) The game was fullscreen only and usually fluctuated around 30fps on the 5770. On the 7950 with Windows I got around 35fps to 45fps. I left the CPU monitor running while the game was going and looked at the chart afterwards. It also appeared to be maxing out 2 cores most of the time.
Not exactly a scientific analysis, as I should've done all of the benchmarking under Windows, but my main motivation was to get 60fps in very specific cases and figured a card with 3x the benchmark scores of my old one would give the results I wanted. It didn't, which made me realize that my next step should probably be replacing the machine. I've had it for 6 and half years now, which is the longest I've used the same desktop so it definitely was a good choice.
The general point of my post was that from a gaming standpoint, I think the 3,1 is old enough that GPU is starting to not be the limiting factor of performance and a card from 2010 to 2012 is probably the best value for general use. Sorry for rambling on.
Register on MacRumors! This sidebar will go away, and you'll see fewer ads.