OWC test lab says the 30" Cinema works on the MDP of their 5870.
Maybe my MDP to Dual-Link DVI Adapter is defective.
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.
So, who has tried their 5870 in a 2008 or 2006 Mac Pro?
- Thread starter ventro
- Start date
- Sort by reaction score
OWC test lab says the 30" Cinema works on the MDP of their 5870.
Maybe my MDP to Dual-Link DVI Adapter is defective.
Here are the Boot ROM versions for the MacPro1,1 Woodcrest I have documented:
MP11.005C.B00 -- boots 5870 and 5770
MP11.005C.B08*-- OWC has this one; they say it boots 5770 but not 5850
MP11.005D.B00 -- boots 5870; will try 5770 Monday
The MP21.007F.B06 listed above looks like the MacPro2,1 Boot ROM version (April 2007).
I think the older spec PCI slot is one reason. For one thing, I didn't check was whether the #1 PCIe slot was running at x16 or x8. Also, the CPU to GPU communication is slower than the newer Mac Pros.
It will be interesting to see if the 5870 is so hamstrung that the 5770 runs just as fast on the 2006 Mac Pro. I'll have an answer Monday when I go back to PowerMax for a second round of testing.
The older PCI-Express 1.0 specification shouldn't be that much of a bottleneck.
I think this article is worth a read. I'm linking to the performance summary page as I think it's of most relevance, but feel free to flick through the other pages.
Notably, they ran the card at 2.0 8x to simulate performance in a 1.0 16x slot, and the overall performance hit is negligible: an average of 1-2%. Current 2.0 GPUs just don't saturate the bandwidth enough that running a 2.0 card in a 1.x slot is almost as comparable to running a 2.0 card in a 2.0 slot.
I think it's definitely worth checking Expansion Slot Utility.app to check whether Slot 1 was running at 16x or lower.
Can't say it enough, but thanks again for all your hard work.
Barefeets,
Instead of trying to locate a MP with MP11.005C.B08, I would try Apple's EFI Update to update the MP11.005C.B00 to MP11.005C.B08. At least that what the article says. http://www.apple.com/support/downloads/macproefifirmwareupdate12.html Since most of us back in the day had already applyed the Update.
After I finish one more round of testing the 5870 and 5770 in the 2006 Mac Pro on Monday, I'll try updating the Boot ROM. Then I give it another try.
Meanwhile, today I'm back on the 6-Core Westmere with a 5770. I'll have results comparing it to the 5870 in an hour or so.
We'll definitely been watching for the results, you guys are doing a great job. Keep up the great work!! =D
Another "taste" of what I'm testing today on the 6-core Westmere...
ET:QW at 2560x1600, 4x Multisampling, High Quality:
5870 = 109 fps
5770 = 59 fps
4870 = 64 fps
GTX 285 = 84 fps
Portal (Full) at 2560x1600, 4x Multisampling, High Quality:
5870 = 127 fps
5770 = 70 fps
4870 = ** fps
GTX 285 = 101 fps
(** Portal and Team Fortress 2 run fine on the 4870 when it's in the 2009 Nehalem but I get kernel panics when I launch Portal or Team Fortress 2 with the 4870 in the 2010 Westmere. All non-Steam games run fine.)
ET:QW at 2560x1600, 4x Multisampling, High Quality:
5870 = 109 fps
5770 = 59 fps
4870 = 64 fps
GTX 285 = 84 fps
Portal (Full) at 2560x1600, 4x Multisampling, High Quality:
5870 = 127 fps
5770 = 70 fps
4870 = ** fps
GTX 285 = 101 fps
(** Portal and Team Fortress 2 run fine on the 4870 when it's in the 2009 Nehalem but I get kernel panics when I launch Portal or Team Fortress 2 with the 4870 in the 2010 Westmere. All non-Steam games run fine.)
I'm an equal opportunity game tester. ;-)
I am constantly seeking to update my Mac game test suite. I added Portal and Team Fortress 2 this year. I have Starcraft 2 but there's no easy way to benchmark it, though I hear the next release will have a benchmarking option. Ditto for BioShock for Mac.
What would you consider a good mix for a Mac gaming test suite?
Huh? ET:QW uses an upgraded version of the Quake 4 engine, which is still a pretty recent engine.
My mistake, I somehow thought it was the Wolfenstein ET. Also, there is no Quake 4 engine. Quake Wars Is based off of id tech 4 (aka Doom 3 engine) which is over half a decade old (6 years this month if I recall correctly).
http://en.wikipedia.org/wiki/Id_Tech_4
ET:QW not only uses a massively upgraded version of the Tech 4 engine, it even uses megatexturing from the unreleased Tech 5 engine.
It's hardly an out of date engine.
Intel even has a version that's ray traced. Looks purty.
Of course there is. Set up a crazy demanding replay (lots of units fighting each other etc etc) and run whichever machine you want through the replay.
haven't really put much thought into it, But I know it wouldn't include Halo, a game almost a decade old.
I stand corrected. Still, haven't seen anyone (besides barefeats) use ET:QW for benchmarking in...years. Though, I suppose you're limited on the Mac side.
The Source engine has been enhanced throughout the years to keep pace with the other engines. Half Life 2, back in the day, did have a benchmarking mode. Dunno if it's been upgraded to use newer effects, and it's possibly the same one Portal used. I just remember doing a lot of benchmarking on it back in the day, because it did a lot of effects that were brand new at the time.
I know HL2: The Lost Coast had a benchmarking mode for the effects that were new when it was relevant (which was just HDR iirc). Though, to my knowledge, the new effects/updates have only been applied to TF2/L4D1/2 (more so 1 than 2).
Can you please let us know if it has been confirmed that a 5870 will not work in a 1,1 Mac Pro with the boot rom MP11.005C.B08?
Just posted this...
SERIOUS FUN: Radeon HD 5870 versus 5770 and three other GPUs (4870, GTX 285, FX 4800):
http://www.barefeats.com/wst10g4.html
(Includes Portal, WoW, X-Plane, Motion, OpenGL Extensions Viewer, and SmallLuxGPU results.)
Now I'm packing up to go test the 5770 on the 2006 Mac Pro at PowerMax.
SERIOUS FUN: Radeon HD 5870 versus 5770 and three other GPUs (4870, GTX 285, FX 4800):
http://www.barefeats.com/wst10g4.html
(Includes Portal, WoW, X-Plane, Motion, OpenGL Extensions Viewer, and SmallLuxGPU results.)
Now I'm packing up to go test the 5770 on the 2006 Mac Pro at PowerMax.
5770 works on PowerMax 2006 Mac Pro
I'm back at the PowerMax lab. I'm running the Radeon HD 5770 on the 2006 Mac Pro. Working fine.
Right now I'm checking out my hypothesis that it is just as fast as the 5870 on the "first gen" Mac Pro.
Let's say the gap has narrowed to the point that the cost of the 5870 is questionable for a 2006 Mac Pro. Here's one example:
X-Plane 9.6.1 at 1920x1200:
5870 = 83 fps
5770 = 80 fps
Full report coming later today.
I'm back at the PowerMax lab. I'm running the Radeon HD 5770 on the 2006 Mac Pro. Working fine.
Right now I'm checking out my hypothesis that it is just as fast as the 5870 on the "first gen" Mac Pro.
Let's say the gap has narrowed to the point that the cost of the 5870 is questionable for a 2006 Mac Pro. Here's one example:
X-Plane 9.6.1 at 1920x1200:
5870 = 83 fps
5770 = 80 fps
Full report coming later today.
Probably a null question,but anyway : Do you happen to have resources to test in on larger resolutions,say normal 30" 2560 x 1600 res?
Many end users start to have the bigger screens,and maybe the larger resolutions might push some bigger differencies between the cards?
Edit : I am a idiot, you allready are testing at the said resolutions...
CARRY ON! NOTHING TO SEE HERE!