PDA

View Full Version : How do YOU guage your machine's speed?


maclamb
Jan 31, 2002, 06:37 PM
How do you tell how fast your machine is?

Running old apps and comparing?
benchmarks?
Count the secs for the app to open???

I truly wonder if we can tell the difference in speed in non-graphic (games, 3d, audio) apps?

A short story of how my end-users could tell machine speed:

Working on an old, distrubuted, light-pen based system at a hospital:
Nurses and other "non-technical" folk could tell machine speed because as they light-penned their way throgh the screens (as they had done millions of times), they would postion their light pen over where they expected the next screen to display their choice.
If they got their pen there before the screen could repaint we would get calls asking "what's happend to the system"
bear in mind we're talking about a .5 - 1.5 second delay at most ;-)

For me? If I can click form one app to the next, through the finder, etc with NO delay - that's good enough...

and you???

eyelikeart
Jan 31, 2002, 07:07 PM
for me.....it's when running photoshop....

how fast it rasterizes.... is pretty processor intensive....

DakotaGuy
Jan 31, 2002, 07:12 PM
The only way I can really tell speed difference between my two machines is how many times an icon has to bounce on the bottom of the dock before an application actually opens. Examples are my iBook opens iTunes in one bounce and my iMac takes two. Appleworks takes two bounces in my iBook, three on my iMac. iPhoto takes 5 bounces on the iBook and 7 on the iMac. iMovie is probably the most memory intensive program I have and it takes a whopping 26 bounces to open on my iMac but only 14 on the iBook. I have 256MB of RAM in the iBook and only 128MB in the iMac...plus the iBook has 200 more Mhz so I suppose those times are fairly representive of the Mhz and RAM. I am not saying this is a scientific way to do it...I just did it for fun to see and in this case Mhz does make a difference. Most of my apps are apple apps and the rest are things like Fire, Limewire, IE, etc... so I cannot comment on the difference in very intense programs like Photoshop, etc.

Beej
Jan 31, 2002, 07:34 PM
Q3A framerates ;)

And Photoshop things.

And App opening times.

And general zippiness in OS X. I don't own a single machine I would call "zippy" in OS X (nor have I ever used one) so it's a good way to judge.

But mainly Q3A framerates :D

748s
Jan 31, 2002, 07:39 PM
after effects renders

Catfish_Man
Jan 31, 2002, 08:11 PM
...my G3 233MHz is at the low end of "useable" (only lags interminably on a very few tasks). The DP800 I tried was "great" (minimal [if any] lag on anything). Also, 26 bounces for iMovie sucks, my G3 233MHz does it in 9-10 with three IE windows and AIM open (that's the first time I've started it in over a week, so it wasn't cached).

Onyxx
Jan 31, 2002, 09:46 PM
final cut pro render speeds at full quality

Cinema 4d xl example files rendered with 3x3 sampling for textures, trans/refraction on, and a resolution of 720x540. I see how long it takes to do a 3 second animation at 30 fps.

I have to say that games are a really crappy way of testing how fast a machine is. tell me which will run faster frame rates, g4 dual 800 with a rage 128, or a dual 800 with a geforce 3? hmmm. me thinks that any sort of "game test" is more than a little bit biased to the capabilities of the graphics card in the machine.

I agree with the photo shop test. A great way of guaging speed cross platform is to to make a photoshop file of about 8x10 inches @ 300 dpi resolution with a slew of actions (and i do mean a slew, about 20-30 should do) let this run and see how long it takes your machine to run it.

animemaster
Jan 31, 2002, 11:55 PM
Probably like most of you, I calculate how fast my machine is by how fast I work in most of the Adobe products I use, and how fast I can render video in programs such as Final Cut Pro, iMovie, Premiere, and Bryce.

Thanks,
-AnimeMaster

GigaWire
Feb 1, 2002, 04:00 AM
take it to the top of a building, drop it, and count one-one thousand, two-one thousand, three-one thous...

shergar
Feb 1, 2002, 04:31 AM
Like most people I guage the speed of my machine on render times. I never render After Effects on my G3 400 powerbook, but I'm quite happy to do so on my G4 400 - even if it means letting it chug away over night.
A much easier way to gauge computer speeds is by their startup times. The g4 takes about a minute and five seconds from pressing the power button to being usable, whereas the powerbook takes almost a minute longer (both using X 10.1)

I have to say - and this worries me slightly - that the more I use OS X the longer it is taking each computer to startup. Anyone else notice this?

Microsoft_Windows_Hater
Feb 1, 2002, 06:54 AM
I don't care about how fast it is...as long as it gets my work done.

I have used a snappy OS X, on a TiBook 667. That was snappy as anything.

About start times, I luckily never have to start my machine as it takes tooo long. But there are things that can make the start time worse such as networking and the fact that the HD can be fragmented.

tcolling
Feb 1, 2002, 01:29 PM
Originally posted by GigaWire
take it to the top of a building, drop it, and count one-one thousand, two-one thousand, three-one thous...

I think my way is better. Roll it down a hill and follow behind with my car, checking the speedometer the whole way.

I am expecting the new G5 sphere to be SOOO much faster!

maclamb
Feb 1, 2002, 01:45 PM
couple of things:

1. Dropping it off the building is not a good test because all things will reach the same terminal velocity after a period of time, so after taking into account wind resistence (potentially minimzed with a round G5 case - see other threads) macs and peecees would fall the same

2. My mac seems to boot faster in osx 10.12 than ever before

3. what do you think of measuring time to load Classic env? - Me thinks that would be a good combination of drive, architecture and CPU???

dantec
Feb 1, 2002, 03:30 PM
Originally posted by Abercrombieboy
The only way I can really tell speed difference between my two machines is how many times an icon has to bounce on the bottom of the dock before an application actually opens. Examples are my iBook opens iTunes in one bounce and my iMac takes two. Appleworks takes two bounces in my iBook, three on my iMac. iPhoto takes 5 bounces on the iBook and 7 on the iMac. iMovie is probably the most memory intensive program I have and it takes a whopping 26 bounces to open on my iMac but only 14 on the iBook. I have 256MB of RAM in the iBook and only 128MB in the iMac...plus the iBook has 200 more Mhz so I suppose those times are fairly representive of the Mhz and RAM. I am not saying this is a scientific way to do it...I just did it for fun to see and in this case Mhz does make a difference. Most of my apps are apple apps and the rest are things like Fire, Limewire, IE, etc... so I cannot comment on the difference in very intense programs like Photoshop, etc.

If your iMac has been running OS X for longer than your iBook, and speed of opening apps, has become sort of sluggish, open your terminal and type this...

sudo update_prebinding -verbose -root /

Make sure you include the slash at the end of the line. This process will take about 15 to 30 minutes depending on your machine, and how many apps your machine has. This bascially rejoins the apps processes, and files. This is usefull because if an app doesn't know right away where the files are, it must search the HD to find them (this was one of the main bottlenecks in 10.0).

dantec
Feb 1, 2002, 03:33 PM
Originally posted by shergar
Like most people I guage the speed of my machine on render times. I never render After Effects on my G3 400 powerbook, but I'm quite happy to do so on my G4 400 - even if it means letting it chug away over night.
A much easier way to gauge computer speeds is by their startup times. The g4 takes about a minute and five seconds from pressing the power button to being usable, whereas the powerbook takes almost a minute longer (both using X 10.1)

I have to say - and this worries me slightly - that the more I use OS X the longer it is taking each computer to startup. Anyone else notice this?

This is because your HD is getting fragmented... Whatever you do... don't optimize your drive with Norton Utilities, it makes OS X even slower! This is because it marks most of the OS X's system files as "other" and scatters them all over the bottom of your drive! Try an update_prebind technique like I used above... You'll just have to wait with the rest of us for an optimizer of OS X!

dantec
Feb 1, 2002, 03:38 PM
Oh yeah...

I like to use Quake to see how my Powermac's GF3 is doing... and to someone else in this thread... the CPU is still responsable. A dual 800 with GF3 can have around 130 fps, where as my 867 with GF3 gets around a maximum of 102!

I use Photoshop, bouncing of dock icons... etc...


And to Gigawire, maclamb & tcolling, how's my Quicksilver gonna roll down the hill? I could see an iMac, but my G4 no...