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

PowerMac G5 performance, and games

Originally posted by 000111one111000
My brother just got his G5 earlier today. It's a 1.8 GHz model with 512 MB RAM.

I forgot to check which video card it has, but I'll check it tomorrow when I go to set up his net connection. I'm assuming it's the fx5200 though.

If you want though, give me a list of things to do, and I'll test em out.

Games are pretty much a no go though, unless I download America's Army. The only games I own are the original Quake 3 Arena and Unreal Tournament.

But any other apps or actions you want me to test, just let me know, and I'll do my best to test it all.

enoch
Click to expand...

Q3:A, UT and AA are all great if you get a chance... also, see if you can find out what the video card is!

thanks ;)

-- james
 
Are the OS X versions of Q3A and UT any good? I haven't tried them in forever, but if I remember correctly, they weren't all that great. Especially the UT one. If I remember though, I'll try them out.

enoch
 
Originally posted by 000111one111000
Are the OS X versions of Q3A and UT any good? I haven't tried them in forever, but if I remember correctly, they weren't all that great. Especially the UT one. If I remember though, I'll try them out.

enoch
Click to expand...

Good point... forgot entirely about that.

Q3:A should be fine, it's UT that'll be the problem. The guy that is porting UT:2003 is also having a stab at UT though in his spare time.

-- james
 
MacWorld tested the G5s with Q3A 1.30b5 so the latest 1.32 should be great since it was much more reliable and quick.

I still don't have any issues (that weren't on Westlake's list) with the UT preview for Mac OS X, but knowing that others are makes me wonder about it running decently on a G5.
 
Originally posted by bousozoku
MacWorld tested the G5s with Q3A 1.30b5 so the latest 1.32 should be great since it was much more reliable and quick.

I still don't have any issues (that weren't on Westlake's list) with the UT preview for Mac OS X, but knowing that others are makes me wonder about it running decently on a G5.
Click to expand...

According to MacWorld:

Quake 3 1.30b5

G5 Dual 2ghz: 268.5 FPS
G5 1.8ghz: 133
G5 1.6ghz: 117.4
G4 Dual 1.42ghz: 134.1
G4 1ghz: 75.6

Anyone else tested anything?

-- james
 
I know there are more getting out there into the channel... anyone else had a chance to put one through it's paces?

thx

-- james
 
xBench results for my brother's G5.

All stock, nothing extra installed yet.



Results 152.69
System Info
Xbench Version 1.1.1
System Version 10.2.7 (G5) (6S80)
Physical RAM 512 MB
Model PowerMac7,2
Processor PowerPC 970 @ 1.80 GHz
L1 Cache 64K (instruction), 32K (data)
L2 Cache 512K @ 1.80 GHz
Bus Frequency 900 MHz
Video Card GeForce FX 5200
Drive Type ST3160023AS
CPU Test 164.23
GCD Loop 100.66 3.93 Mops/sec
Floating Point Basic 266.84 964.99 Mflop/sec
AltiVec Basic 123.42 3.59 Gflop/sec
vecLib FFT 171.43 2.66 Gflop/sec
Floating Point Library 353.59 14.15 Mops/sec
Thread Test 103.73
Computation 73.76 590.58 Kops/sec, 4 threads
Lock Contention 174.69 2.19 Mlocks/sec, 4 threads
Memory Test 259.93
System 255.25
Allocate 468.40 315.85 Kalloc/sec
Fill 175.17 1394.36 MB/sec
Copy 255.80 1278.99 MB/sec
Stream 264.79
Copy 226.28 1654.09 MB/sec [G5]
Scale 230.67 1702.31 MB/sec [G5]
Add 304.85 1951.07 MB/sec [G5]
Triad 325.55 1989.14 MB/sec [G5]
Quartz Graphics Test 181.96
Line 203.46 5.18 Klines/sec [50% alpha]
Rectangle 188.96 13.29 Krects/sec [50% alpha]
Circle 202.71 4.67 Kcircles/sec [50% alpha]
Bezier 166.11 1.80 Kbeziers/sec [50% alpha]
Text 158.27 2.58 Kchars/sec
OpenGL Graphics Test 211.24
Spinning Squares 211.24 147.82 frames/sec
User Interface Test 172.62
Elements 172.62 55.52 refresh/sec
Disk Test 110.83
Sequential 115.15
Uncached Write 146.31 58.24 MB/sec [4K blocks]
Uncached Write 137.10 53.51 MB/sec [256K blocks]
Uncached Read 81.64 12.92 MB/sec [4K blocks]
Uncached Read 119.62 48.33 MB/sec [256K blocks]
Random 106.83
Uncached Write 103.53 1.48 MB/sec [4K blocks]
Uncached Write 107.37 24.22 MB/sec [256K blocks]
Uncached Read 100.64 0.66 MB/sec [4K blocks]
Uncached Read 117.19 24.12 MB/sec [256K blocks]



Tomorrow, I'm gonna try the UT 2003 demo on it. I installed America's Army, but haven't had a chance to try it out too much yet.


if you want to know anything specific, besides games, let me know, and I'll see what i can do.

enoch
 
Originally posted by 000111one111000

Tomorrow, I'm gonna try the UT 2003 demo on it. I installed America's Army, but haven't had a chance to try it out too much yet.


if you want to know anything specific, besides games, let me know, and I'll see what i can do.

enoch [/B]
Click to expand...

hey,

thanks! :)

he's got the std graphics card I take it?

-- james
 
ok, I just installed and ran the UT 2003 Demo.

With all settings set to "highest" and a display size of 800x600, I got a constant 80-90 fps. When things would get really busy, it'd drop down to 30-40 fps for only a second or two, and then shoot back up to 80-90 fps.

Remember, this is the stock config. 512 MB RAM, GeForceFX card, etc.

I'm sure if you pop one of the Radeons in and boost the RAM, it'd fly through games.

If you want to know anything else, just tell me. :)

enoch
 
Yeah ! i saw those tests ! I was surprised to see that the G5 is not as fast as I tought ! I just bought a Single G4 1.25 for 1649 canadian with the radeon 9000. My question is : How faster is a dual G4 compared to a single G4 (in games, apps, etc) when the app or game is not optimized for mp's machines ! nothing ? 10 % ? is a CPU processing the sound, when another doing somthing else ? For those reasons, i tought It was a better option for me to buy a single G4 (since i dont use any mp's apps, or rarely) and to put the extra 500 $ canadian on a Gfx card (when the 9800 will drop a bit) What do you think ?
 
No, a dual-processor machine is always quite a bit faster in nearly anything. In UT 2k3 for example, the second processor deals with the sound, while the first does everything else... but EVERY OS X-native app takes advantage of MP, and therefore every app benefits from dual processors tremendously.

There's a reason a single 450 is considered horribly outdated, while a dual 450 still isn't. Duals are more current for much longer, as well as being faster machines.

Shoulda gotten a dual... ;)
 
I think i'll do !
Is it possible to change an order once it's done ? I only ordered yesterday !
I just call and tell them to add a CPU to my single G4 ?

thanks for the advices
 
Originally posted by jamesa
According to MacWorld:

Quake 3 1.30b5

G5 Dual 2ghz: 268.5 FPS
G5 1.8ghz: 133
G5 1.6ghz: 117.4
G4 Dual 1.42ghz: 134.1
G4 1ghz: 75.6

Anyone else tested anything?

-- james
Click to expand...

Well, although it's a big leap, that seems not so good in comparison to the newest Intel and AMD machines:

Quake3 - Demo1

Pentium 4EE 3,4 GHz: 414,4 fps
Pentium 4EE 3,2 GHz: 401,6 fps
Ahlon64 FX 51 2,2 GHz: 349,1 fps
Pentium 4 3,2 GHz: 390 fps
 
Originally posted by Veldek
Well, although it's a big leap, that seems not so good in comparison to the newest Intel and AMD machines:

Quake3 - Demo1

Pentium 4EE 3,4 GHz: 414,4 fps
Pentium 4EE 3,2 GHz: 401,6 fps
Ahlon64 FX 51 2,2 GHz: 349,1 fps
Pentium 4 3,2 GHz: 390 fps
Click to expand...

I always have to question gaming benchmarks. Are gaming tests consistent? How many objects are on the screen and how fast are they moving? Everyone always questions Photoshop and other apps' benchmarks because they use different files and sizes. Why is this not the case with gaming benchmarks. I haven't seen any indication that there is a standard set of tests that these games use for a benchmark. A game "benchmark" defies the very definition of the word if there is nothing consistent to compare them with. Please show me some standards!
 
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
Back