Its basically a developers toolkit. It has a few other apps as well that let you change the performance of your system to simulate different conditions.
I can run single or dual processor, 0 or 256k of LII cache, 0 or 1MB of LIII cache. These digures naturally depend on your system. There is also a 'NAP' mode that idles the processor when its not used. Operating temps really go down (as does fan noise) but it causes random freezes in most machines that aren't designed to do it (eg the MDD that I have). I think the G5s use it.
Any processor intensive function will vary greatly depending on the ammounts of cache. Add a little more cache to the same processor and results increase more than a small boost in Mhz.
On a different note, I'm curious about some of the results. 3ghz windows machines usually come in around 60 seconds. Yet some with sub 3ghz are reporting times of around 50s? This right? What gives?
p.14 #4 · Let us testing Photoshop speed in our system
49.7 seconds after shutting down all the bs running in the background. Not much of a performance boost, but d*** PS 6.0 is slow... i was so disappointed with the machine after I ran the test on 6.0 before I got CS today.... *whew*
*steve
ps the machine runs half life 2 at the highest resolution and all the graphics options on the highest settings *perfectly* Gotta have some fun sometime!
p.14 #8 · Let us testing Photoshop speed in our system
45.7s
Dual AMD Athlon MP 2400's ( overclocked to 2800's @2.10 GHz)
1 GB of DDR 3000 Ram (@138 MHz and Cas 2)
Win XP Pro w/ SP2
PS CS
Radeon 9500
3x 80GB WD HDD 7200RPM 8MB cache in RAID 0
As others have mentioned here previously, this is strictly a cpu test as both of mine were maxed out at 100% utilization during the test and probably no scratch disk usage at all.
Very happy with the desktop, since I just built it for only
1200...holds its own with the single processor machines..
but those dual macs make me think .....is that treason?
p.14 #13 · Let us testing Photoshop speed in our system
28s
Power Mac G5 64-bit Dual 2.5GHZ liquid cooled
512K L2 cache/processor
4 GB DDR400 SDRAM
ATI Radeon 9600 XT
23" Titanium Apple Cinema Display HD
Photoshop CS
This was with WoW running as well as Final Cut Pro HD because I couldn't be bothered to shut them down. It would have been under 20 seconds with nothing else running.
Although this should not be surprising. According to a series of benchmarks the dual 2.5 G5 ran photoshop CS 98% faster than a Dell Precision 650 Dual 3.2 Xeon.
p.14 #16 · Let us testing Photoshop speed in our system
I'm not sure how that changes the fact that with 3 labor intense apps running the dual 2.5 G5 runs the test super fast.
With those apps turned off it runs the test in just over 18s
Of course companies will always show there best face in there tests but it also doesnt mean that the tests were fabricated. Nor does it change the fact that 18s-28s is fully 20-30 seconds faster than my Dell dual Xeon 3.0 sitting 10 feet away.
p.14 #17 · Let us testing Photoshop speed in our system
cullwynn wrote:
Nor does it change the fact that 18s-28s is fully 20-30 seconds faster than my Dell dual Xeon 3.0 sitting 10 feet away.
If your dual Xeon is running this test at 40-60 seconds, there must be something wrong. These other Xeons perform much better. Yes the PMG5 shows remarkable performance here, but your Xeon is very much an outlier when you look at similar systems.
p.14 #18 · Let us testing Photoshop speed in our system
My dual Xeon rig runs the test at approx 35s. My only point with all of this is that the multi-tasking performance is really untouchable. The Xeon is is 7s slower than the PMG5 with Final Cut Pro HD running and World of Warcraft going in full swing as well.....that alone is simply crazy. It is also 8 seconds faster than the Xeon while FCPHD is running and rendering a 3G network TV bumper.
Running a PS CS while running anything else on the PC is a little touch and go. If you try to run anything that is activly using the hard drive then your screwed.
I mean lets put this in perspective. We are talking about computers that run $2500-$4500. When you throw in the monitor you can esily add another $700-$3000. My PMG5 setup with the 23" Titanium HD display cost somewhere around 6k.
All of these computers perform amazingly well, and they should. A difference of 6-8 seconds does not mean much to me without a true baseline for the test computers. But 15-30 seconds starts to get up there (as some of the dual Xeon users reported....27s, 30.8s, 44s) The next question really to be asked is how much those machines cost, and what is the performnce to price ratio :-) Also one of my biggest concerns has always been how well a compuer can multi-task. Until the HT tech, the answer for a windows box since NT has been "pretty well". I love the G5's because you can get them to do so much at one time.
I am not trying to argue anything, the numbers are simply that...numbers. I like OS X and I like Windows XP Pro. If I didn't work in the TV animation industry and could only afford to buy one of the machines I would get the Mac, but that is a personal prefrence. I like having the abilty to have 6 programs running well all at once.
I would like to know how much those 5-10-20 second increases cost you though...that should be the next big pole. :-)
p.14 #19 · Let us testing Photoshop speed in our system
cullwynn wrote:
If I didn't work in the TV animation industry ...
Need a copy of Boris Red 3GL?
Point taken about multitasking. That is amazing how fast it runs with the FCPHD render going. Just wait until OSX and FCP go fully 64bit.
Another idea to test...scratch drive importance. This test does not make a difference if you run it off the sytem or scratch disk. Is there a way to run a test that introduced the scratch disk speed?
I'd aslo like to see an action built that uses layers and history states. That should add RAM to the equation. Something that gets the dual G5s up around the 60 second mark. The longer the test takes the more it seperates the contenders.
It's interesting that my wife's iMac G5 will significantly out-perform this. I would be running a Mac myself if it were not for the fact that I need the PC for my "real" work. I would hate to give up the 22" monitor though. I guess that means having to shell out the big bucks for one of those Apple 23" (or 30" *drool*) Cinema Displays. If only I were made of money
Gotta have some fun sometime!
