> Hi. I've got an iMac G5, and I'm unsure whether or not it's performing
> up to scratch. It often runs so slowly that I suspect hardware problems.
>> Hi. I've got an iMac G5, and I'm unsure whether or not it's performing
>> up to scratch. It often runs so slowly that I suspect hardware problems.
>
> Have you tried an Archive and Install?
I haven't, but was considering it. I was worried that the problem would
then be "fixed" without me working out what was causing it.
> I think if something is slowing your system down then a benchmark may
> be slowed down by it too, depending on what operation is being
> performed. There are a few benchmark programs out there for Mac (again,
> search Apple or VersionTracker)
Great idea; I didn't think to look on VT. There were quite a few hits,
but I found a great app called Xbench that allows you to compare your
computer's performance to similar models.
http://www.xbench.com/
You can see the results of everybody's benchmarks here:
http://db.xbench.com/
I ran the tests with 512MB of RAM in (search for "clown" from the
second link) and scored 76.87 overall.
I then took the lid off, slapped in a couple of 1GB sticks, and ran the
tests again. You can see (in detail) how things changed on the site by
getting a direct comparison between two different test runs, but
basically, the overall stats went up to just over 80. There seems to be
some random deviation in there, but that's to be expected.
The most telling thing of all is that the usability of the system has
gone up enormously. I ran all the apps I had in the Applications folder
and only used up half my RAM (maybe I should have opened some files).
Even though the disk was getting hammered and the dock was full of
bouncing icons the desktop was still responsive; it never felt like
that with 512MB.
I'm very very pleased by the results of the upgrade; it's turned it
into a totally different computer. What surprises me is that the Xbench
stats indicate that the previous RAM wasn't faulty, and yet my iMac
performed so badly compared to what I had been expecting.
Thanks for the pointer.
Graham
Jonny Morris - 27 May 2006 15:18 GMT
>>> Hi. I've got an iMac G5, and I'm unsure whether or not it's
>>> performing up to scratch. It often runs so slowly that I suspect
>>> hardware problems.
> I'm very very pleased by the results of the upgrade; it's turned it into
> a totally different computer. What surprises me is that the Xbench stats
> indicate that the previous RAM wasn't faulty, and yet my iMac performed
> so badly compared to what I had been expecting.
Maybe the G5 uses more system RAM *shrug* I've run OSX Panther quite
happily on a G4 with 512MB, it only slowed down when I started doing
Photoshop etc. In general though, the advice for OSX is max out with as
much RAM as you can afford!
Adrian - 27 May 2006 16:27 GMT
> I'm very very pleased by the results of the upgrade; it's turned it
> into a totally different computer. What surprises me is that the Xbench
> stats indicate that the previous RAM wasn't faulty, and yet my iMac
> performed so badly compared to what I had been expecting.
Had you ever run the hardware tests on the Install discs. These give the
RAM a much more thorough test over a period of time. However, I believe
that some RAM faults don't even show up with that. My bet is that you
have successfully removed faulty RAM, as the extent of the improvements
you describe would not be quite as dramatic (aside from the multiple
apps test you did, where you would expect a significant difference.) I
have run a G5 PowerMac with 512 and it did jut fine ... of course now
that it has 3.5GB, ahem, it does even better!

Signature
Adrian
Graham Ashton - 27 May 2006 18:25 GMT
>> I'm very very pleased by the results of the upgrade; it's turned it
>> into a totally different computer. What surprises me is that the Xbench
[quoted text clipped - 3 lines]
> Had you ever run the hardware tests on the Install discs. These give the
> RAM a much more thorough test over a period of time.
I have, yes. That's what the guys at the local Genius bar recommended.
I was expecting the RAM to come up faulty, but it all passed.
> However, I believe that some RAM faults don't even show up with that. My
> bet is that you have successfully removed faulty RAM
I suspect you might be right. The guy I talked to at the Genius bar
said it didn't get all RAM faults, but did pick up nearly all of them.
Perhaps I should take the stick back and get them to test it...
Graham
Adrian - 27 May 2006 19:50 GMT
> I suspect you might be right. The guy I talked to at the Genius bar
> said it didn't get all RAM faults, but did pick up nearly all of them.
>
> Perhaps I should take the stick back and get them to test it...
I would be surprised if they tested any differently to what you have
already done ... they'll tell you it's OK ...

Signature
Adrian