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

aramosc

macrumors regular
Original poster
May 4, 2011
225
0
San Diego, CA
hi everyone. I have a Mac Pro 3,1 2.8ghz 8-core with 8 gb ram and stock 2600 graphics and stock 320gb wd hdd. and I ran xbench and got a user interface test in the 15 point range is there anything wrong with the machine? is there anything I can do to fix it?
 
look at my score and the drive I use. I get 444.91 with a new intel 320 series ssd it uses 200gb out of 300gb. give me 10 minutes I will test with a different drive. okay second test is a superduper exact clone of the ssd.


both scores are much better then yours in fact the non ssd is better then the ssd. you need to do a few tests if you have a second boot drive on a different hdd test it. if not use superduper make a clone and test it. if you score 15 it is not the hdd
 

Attachments

  • Screen shot 2011-05-10 at 8.42.31 PM.jpg
    Screen shot 2011-05-10 at 8.42.31 PM.jpg
    225.2 KB · Views: 174
  • Screen shot 2011-05-10 at 8.50.52 PM.jpg
    Screen shot 2011-05-10 at 8.50.52 PM.jpg
    223.1 KB · Views: 142
Last edited:
I don't thinnk it's the HDD. there is a separate category for HDD performance. But yeah the difference from my xbench and geekbench is just abysmal in xbench i'm in the G5 level and geekbench it's high up there..
 
I agree, but hdd is the easiest to eliminate. if a clone copy gives you terrible scores then you can remove it.

You have a small hdd 320gb if it is filled beyond 240gb it may be the problem. Actually I am curious as to a score that low.

Do you have a lot of add on units via usb firewire two or thee lcd screens. Do you have a wireless keyboard and mouse?

User interface score may be the speed of your keyboard connection. I use a wired key board. I do have an imac with a wireless keyboard. If you wait a bit I will go test it. Here is the test with my 2009 iMac it is slower then the 2010 macpro 330 vs 480. The hdd is 1tb and is half full. this keyboard is wireless. I still suspect that a very full hdd may be your problem. I am thinking wired or wireless keyboard is not the problem.

I also know that geekbench basically does not care what your hdd is and xbench uses a lot of hdd in its final scores.

I guess I should say geekbench weights its final score with little emphasis on the hdd.

Xbench uses a lot of emphasis on your hdd in its final score.
 

Attachments

  • Screen shot 2011-05-11 at 8.31.37 AM.png
    Screen shot 2011-05-11 at 8.31.37 AM.png
    687.8 KB · Views: 136
Last edited:
is there anything wrong with the machine?

Is there? I mean, it's your machine, is it working well?

Your actual experience with the machine trumps any artificial benchmark. Last time I checked, XBench was widely held to be useless, and that was at least five years ago, so maybe things have changed.

If you really want to measure the size of your ...user-interface score, why not run the same version of XBench on a friend's Mac, compare the scores and your impression of the relative snappiness of the two computers to tell whether it's XBench or your Mac.
 
I
Your actual experience with the machine trumps any artificial benchmark. Last time I checked, XBench was widely held to be useless, and that was at least five years ago, so maybe things have changed.

this.





don't try to fix something that isnt broken.
 
XBench is like Windows Experience Index. Totally useless and unless you have an SSD or multi HDD RAID the HD test will bring your score waaaaay down. Also if you run the test multiple times you will see slight to major variations. It sucks.
 
yeah but I did two tests and the second one was with a 2tb standard hdd filled with 200gb of info score was 400 and change. Also on an iMac with a 1tb score was 330 he is getting a 15. he really should clone his 320gb hdd to a new 1tb hdd like this one


http://www.newegg.com/Product/Product.aspx?Item=N82E16822152185&Tpk=samsung f3


and retest using it as the boot drive. maybe his geekbench will go from 12000 to 13000. a 15 is a lowww number and indicates something is off.
 
I also failed to ntion that I have 3 more 1 tb drives that are pretty empty. And the 320 gb has 140 gigs free the performance of the drives on the tests is low however the user interface test went uP to the 300s so I'm happy with that the machine overall feels fast so I think it may have been that I needed to reboot and clear all the caches.
 
His geekbench will go nowhere as it does not test the HD's. Memory and CPU only.

the geekbench score does not use much from the hdd for its scores. but if there is a big hdd problem it will lower geekbench. He has done some maintenance and he upped his interface from 15 to 30. I still suspect a bigger better hdd would help him a bit.
 
I am now pulling almost 400 for the interface test.. the HDD is still lagging but I guess that is because my hard drives are not top of the line.. but now that I ran Onyx it feels super fast it boots very fast and applications open before they can bounce once so I'm happy now. I really don't care about the number but I just didn't want to have to wait forever for applications.. but now score is up and the mac pro is healthy! thanks to everyone for helping me out!
 
well he boosted it from 15 to 400 so whatever he did it helped.


Myself I run geekbench
rember
xbench
and aja.

Like you say geekbench does not push hdd scores xbench does and none of them are full examples of real life.
 
Rember and Aja are great tools to have. Cinebench and LuxMark are fairly ok for us Mac benchers with miserable tools. Don't run Geekbench or Xbench unless I have a reason. Actually I don't run any of them unless I get new kit. FWIW my 6-core gets 537.10 on that interface test. 600GB Velociraptor and 5870.
 
Register on MacRumors! This sidebar will go away, and you'll see fewer ads.