Sidebar Sidebar
Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.

OpenMark results on PowerPC Macs

zen.state

zen.state

macrumors 68020
Original poster
Mar 13, 2005
2,181
8
I am very interested to see how different PowerPC Macs perform with OpenMark. It provides very accurate and consistent results and will help us all compare the GPU and overall graphics throughput of different Mac/GPU combos.

Download OpenMark here: http://mac.majorgeeks.com/download5008.html

Please explain what Mac, CPU and OS you have and also take a screen capture of the results window when done. The results window will say the GPU. I won't ask for a certain resolution as the mixed ones will also tell a story.

Once you launch it go to the OpenMark menu and select "Run Benchmark" to start the test.

I will start off:

PowerMac G4 Sawtooth
G4 1.8GHz 7448 running 10.5.8
Geforce 6200 256MB

unledhg.jpg
 
Last edited:
zen.state

zen.state

macrumors 68020
Original poster
Mar 13, 2005
2,181
8
You didn't run the test as there is no score at the top.

Edit: I see you fixed it now. :)
 
Last edited:
N

Nova77

Guest
May 12, 2011
337
1
zen.state said:
Are you saying a Geforce 6200 is a "much better video card" than the Radeon 9700? It isn't. A 9700 would even be faster than a GeForce 6600. Even if it is the mobility GPU. The later PowerBook G4's had very good graphic capabilities and you can see in your test.
Click to expand...

You are wrong. The Mobility edition of the Radeon 9700 is a little worse than the Radeon 9000 Pro , but does support coreimage, so it gives a little plus. So overall, both cards bench almost the same. Check out http://www.videocardbenchmark.net/ if you are interested, as I don't own the 9000 pro anymore.

A "real" Radeon 9700 would be a good card, but the mobility edition just sucks by 2005's standards. As I mentionned before, the powerbook g4 1.67 or 1.5 (not the latest model with faster RAM) feels just like a dual 1 ghz MDD with 1 GB RAM, and an ATI 9000 pro. Also, specs are VERY similar. The G4 MDD is the powerbook G4's closest relative when it comes to desktops. Other G4 MDD and powerbook G4s have mentionned this before, but trust me I've used both quite a lot to know what I'm talking about.

So yeah, your card is better, and still my laptop's benchmark running on battery comes close to your score.

Since the 9700 mobility benches like an 9000 pro desktop one (I know what I'm talking about, since I owned both), it means a stock G4 MDD would be as good GPU wise as your upgraded sawtooth. (edit: that part is related to the previous thread)


So after all... maybe 2X vs 4X AGP does make a difference...
...or else you are saying your card, one of the geforce 6 series, is worse than an ati 9000 pro? sounds weird to me...
 
Last edited:
zen.state

zen.state

macrumors 68020
Original poster
Mar 13, 2005
2,181
8
The Geforce 6200 actually has hardware rendering unlike the Radeon 9000/9200/8500 which are all based on the same basic GPU. So yes a 6200 is better than all those.

Before I had the 6200 in my Sawtooth I had a Radeon 7500 32MB with all the other hardware the same and it scored about 1200. The 6200 is more than 4x that score so it does help a lot.

Don't forget also that your PowerBook is running Tiger which has a faster GUI than Leopard on the same hardware so a Tiger system will score higher. It isn't a fair comparison at all with a different OS.
 
N

Nova77

Guest
May 12, 2011
337
1
zen.state said:
The Geforce 6200 actually has hardware rendering unlike the Radeon 9000/9200/8500 which are all based on the same basic GPU. So yes a 6200 is better than all those.

Before I had the 6200 in my Sawtooth I had a Radeon 7500 32MB with all the other hardware the same and it scored about 1200. The 6200 is more than 4x that score so it does help a lot.

Don't forget also that your PowerBook is running Tiger which has a faster GUI than Leopard on the same hardware so a Tiger system will score higher. It isn't a fair comparison at all with a different OS.
Click to expand...

I can run it again using 10.5.8, no problem. I don't expect much of a difference, since I was on battery for the benchmark.
 
zen.state

zen.state

macrumors 68020
Original poster
Mar 13, 2005
2,181
8
Lets see it on the power adapter and see if there is a difference. Why run a benchmark on the battery?
 
N

Nova77

Guest
May 12, 2011
337
1
zen.state said:
Lets see it on the power adapter and see if there is a difference. Why run a benchmark on the battery?
Click to expand...

I ran it on the passenger's seat while driving in my car... lol Should not have that much of a difference under tiger since I was on "best performance" battery mode. Maybe just a little less power...

Last bench I posted is with the power adapter under 10.5.8. Result hasn't changed a lot, so what I was saying earlier still means something.
 
zen.state

zen.state

macrumors 68020
Original poster
Mar 13, 2005
2,181
8
Nova77 said:
I ran it on the passenger's seat while driving in my car... lol
Last bench I posted is with the power adapter under 10.5.8. Result hasn't changed a lot, so what I was saying earlier still means something.
Click to expand...

Plug the adapter into the nearest tree. ;)
 
N

Nova77

Guest
May 12, 2011
337
1
zen.state said:
Plug the adapter into the nearest tree. ;)
Click to expand...

Same result with Tiger and adapter plugged-in. 5018. Feel too lazy to post screenshot, as it is the same score as the one in 10.5.8 (scroll up...).

Edit: Since I was suspicious about the bench giving the same number 3 times, I tried it with adapter unplugged and "save energy setting". Score sucked bad.... about 1200.
 
zen.state

zen.state

macrumors 68020
Original poster
Mar 13, 2005
2,181
8
Nova77 said:
Same result with Tiger and adapter plugged-in. 5018. Feel too lazy to post screenshot, as it is the same score as the one in 10.5.8 (scroll up...).
Click to expand...

So at least for your PB vs. my G4 it looks like my slightly better GPU makes up for the 2x AGP vs. it's 4x. Your PB has more GPU bus throughput but my GPU has more power.

It's stuff like this I love. Comparing the little subtleties of different hardware in different situations. Love it!.
 
N

Nova77

Guest
May 12, 2011
337
1
zen.state said:
So at least for your PB vs. my G4 it looks like my slightly better GPU makes up for the 2x AGP vs. it's 4x. Your PB has more GPU bus throughput but my GPU has more power.

It's stuff like this I love. Comparing the little subtleties of different hardware in different situations. Love it!.
Click to expand...

Yep! Its cool. And we've proved that AGP bus speed does make a difference. So 4X is better than 2X, but as you said your card is good enough to compensate for the lower AGP bus throughput.
Your sawtooth surely beats any stock MDDs, even though its older technology.

But for high specs AGP cards such as my 7800 GS or the too-hard-to-find 7800 GT, better stay away from AGP 2X imo, because although by putting it into AGP 2X it *might* get close to a Radeon 9800 under AGP 4X, you could get so much more power.

I would be curious to see how AGP 8x from the G5's compare to AGP 4x.
 
zen.state

zen.state

macrumors 68020
Original poster
Mar 13, 2005
2,181
8
Nova77 said:
Yep! Its cool. And we've proved that AGP bus speed does make a difference. So 4X is better than 2X, but as you said your card is good enough to compensate for the lower AGP bus throughput.
Your sawtooth surely beats any stock MDDs, even though its older technology.

But for high specs AGP cards such as my 7800 GS or the too-hard-to-find 7800 GT, better stay away from AGP 2X imo, because although by putting it into AGP 2X it *might* get close to a Radeon 9800 under AGP 4X, you could get so much more power.

I would be curious to see how AGP 8x from the G5's compare to AGP 4x.
Click to expand...

Look how close our scores are to Nilscollection's iMac G5 with x600. It would be AGP 8x. It's only about 20% higher than us (6308). Thats pretty good because an x600 would smoke both our GPU's and it's 8x.
 
cocacolakid

cocacolakid

macrumors 65816
Dec 18, 2010
1,108
20
Chicago
I was curious what it would be on my GE 400mhz G4, but it OpenMark crashes everytime I try to run a benchmark.

400mhz G4 PowerMac
1gb RAM
GeForce 4 MX 64mb video
 
SuperJudge

SuperJudge

macrumors 6502
Apr 2, 2008
449
5
The Triangle, NC
Proof that buses are not necessarily the limiting factor:

PowerMac 11,2 (2.3GHz DC G5, PCIe), 4GB RAM, GeForce 6600 w/ 256MB VRAM
 

Attachments

  • Picture 1.png
    Picture 1.png
    57.1 KB · Views: 265
zen.state

zen.state

macrumors 68020
Original poster
Mar 13, 2005
2,181
8
SuperJudge said:
Proof that buses are not necessarily the limiting factor:

PowerMac 11,2 (2.3GHz DC G5, PCIe), 4GB RAM, GeForce 6600 w/ 256MB VRAM
Click to expand...

Yours clearly shows that this test is truly about GPU power. Very interesting. Right about the same as my G4 and the 2x 1.67GHz PB G4's. Yet look how much lower the PB 1.5GHz with Geforce 5200 is at only 1411. The 6600 in G5 towers was the 6600LE to be exact right?

Your Mac would certainly perform better than Nova's MDD yet his GPU score is almost 15000. Shows how powerful those 7800 really are and reinforces how GPU dependent this test really is.

Love the thread results so far!
 
SuperJudge

SuperJudge

macrumors 6502
Apr 2, 2008
449
5
The Triangle, NC
zen.state said:
Yours clearly shows that this test is truly about GPU power. Very interesting. Right about the same as my G4 and the 2x 1.67GHz PB G4's. The 6600 in G5 towers was the 6600LE to be exact right?

Your Mac would certainly perform better than Nova's MDD yet his GPU score is almost 15000. Shows how powerful those 7800 really are and reinforces how GPU dependent this test really is.

Love the thread results so far!
Click to expand...

It's the plain 6600 in the one that I've got. The 6600LE was in the 2.0GHz DC models. The really interesting thing to me about this test that it appears to be fairly resolution independent. I just ran it again at 1280x1024 and it performed a little worse, actually. (See attached.) Those 7800s are definitely pretty monstrous and now I kinda want get one for my G5 here.

But yeah, I think that talk of bus bandwidth is fairly irrelevant. It's all about the GPU and whether or not it can actually saturate said bus. I doubt most consumer cards even come close.

This is the most interesting thread in quite a while. Moar results, plz! :D
 

Attachments

  • Picture 2.png
    Picture 2.png
    59.1 KB · Views: 219
zen.state

zen.state

macrumors 68020
Original poster
Mar 13, 2005
2,181
8
It is a little sensitive to resolution it seems yes. My original here is at only 1280 so here it is at 1680 and 1920 also. I always run my 24" LCD at the optimal 1920 so I may as well show it for a more real world result.

@1680x1050
picture1n.jpg


@1920x1200
picture2eq.jpg


And even with an Intel C2D MacBook I can't beat my G4 GPU because the MB has the crappy integrated GMA 950.

MacBook C2D 2 GHz running 10.6.8
GMA 950 64MB
unledsvh.jpg
 
SuperJudge

SuperJudge

macrumors 6502
Apr 2, 2008
449
5
The Triangle, NC
zen.state said:
And even with an Intel C2D MacBook I can't beat my G4 GPU because the MB has the crappy integrated GMA 950.

MacBook C2D 2 GHz running 10.6.8
GMA 950 64MB
Click to expand...

Ouch. I have some pretty painful memories of attempting to game with the GMA 915 on my old ThinkPad R52. Does the GMA 950 even support Core Image?
 
You must log in or register to reply here.
Register on MacRumors! This sidebar will go away, and you'll see fewer ads.
Top Bottom