[PixelPete17]

Hey gang,

Am about to run some tests on G5 1.6 with 512MB RAM. I don't have a stick of 512 PC2700 unfortunately, so I can't up the memory...but I will be USING PS 7 with G5 update as I conduct some general tests on about 25MB files.

Tune in around four hours later (I have to drive and get back), and I should be able to put alot of our questions to rest.

See ya soon,

Pixel Pete

PS - I MAY be able to drudge up a stick of 512 now that I think about it...my Athlon system has one 2100 and one 2700 speed 512 DIMM now that it just dawned on me.......but I can't remember if I actually bought the 2700 in the end. Anyway, I'll do what I can!

[:XI:]

Does the RAM need to be installed in pairs in the single cpu machines?

[idyll]

Whoa.. you got a G5?!

[phobos]

Can you please create a series of actions on some small pictures? like 1 or 2 MB. You can blow it up to 50 or even 100MB with image size and create some decent benchmarks.
The small files will help us download them easier and use the same actions you ran to our own G4 machines and see how it compares to the new G5.Some numbers won't mean a thing if we don't have some sort of comparison.

Thank you very much for your time on taking the benchmarks and pleasing all of us greedy people!!
Contact me if you want space for the actions and the pictures to be uploaded

[Eug]

Originally posted by :XI::
Does the RAM need to be installed in pairs in the single cpu machines?

Good point. Yes it does.

I'm told it won't work on the G5 at all if you don't. Now, I do know it works on some PCs though (but at half the speed, since you lose dual channel functionality).

P.S. Try running before and after benches with the G4 version and the G5 update if you can.

[sleepyrenderer]

Just use the 50MB version of the PS7Bench script.

Be sure to set History States to 1 though or your results won't be comparable to results previously posted to the net (it will also keep it from using more than your 512MB of ram).

[1stunna]

hey pete. (i was at the uofa).

i dont think that the 1.6 need to have ram put in in pairs. isnt there only one bank?

[jrod7350]

Oh man . . .

This is what I've been waiting for.

Please be fast, be fast . . .

And yah, use the 50MB files, there's a good comparsion with other Pentium/Athlon systems on Ace's Hardware that are using the 50MB files with 1 History state and PSBench7.

[PixelPete17]

OK guys,

here's some numbers from test that you will have to replicate on your own G4s on your own time in order to create a table of G4 vs. G5 data.

The machine only had 256 RAM, it DID have the G5 optimizer installed on it (as I was standing right there when I had the sales guy do it).

I only have time to post some of the numbers I have, but overall, I'd say I can't really form a concrete opinion on this machine until I see it running with a 768MB or more RAM. The 256 was fine for everyday tasks, OS X felt crisp, and I'm sure Panther will be a beauty of a smooth OS once it's out.

Here's the numbers, if you hae questions just post em, and I'll do what I can to answer by tomorrow. I'm going out tonight (after dinner), so I'll get back to you when I can.

Here ya go:

25.2 MB file (digital photograph if you care) PS 7.01 with G5 update:

Gaussian Blur of 57.6 = 3 seconds
GB of 100 setting = 3.3 seconds

Unsharp Mask 200 amount .7 radius 2 threshold = less than 2 seconds

RGB to CMYK = 4 seconds

Craquelure filter (settings 9 and 3 and 9 respectively) = 15 seconds

next I duplicated my background layer and added two text layers...

Reduce file to 1/2 size (12.1 MB est) = 2 seconds

Resize to 77MB file = 27 seconds

On that 77MB file I ran channel mixer to make a greyscale effect (don't know why, I just use channel mixer alot so I figured "why not") = instantaneous

77MB file rotate 90 degrees = 20 seconds
77MB file rotate 77 degrees (via arbitrary setting) = 5 seconds (have no clue why this was so much faster)

26 MB file rotate 90 degrees = 2 seconds

Liquefy (after about twenty brush strokes on 26 MB file) = 35 seconds

YES, I know some of these tests have no direct usage for benchmarks, but hey, I have the numbers so here ya go...

PS7 Advanced Benchmark (but only with 256 MB RAM remember)

file creation = 15 seconds
rotate 90 = 12 seconds
Gaussian Blur = 10 seconds
USM = 8 seconds for first test 13 seconds for the second (different settings) USM test
Despeckle = 12 seconds

OK, I've got a few more but I have to leave to go eat. Personally, I'm going to hold out for a Dualie or even rev B, but the machine itself is whisper quiet on the salesfloor (but room has various levels of ambient noise), and the machine has great machining and fit and finish. It's really nice if you like industrial minimalism.

I won't be going back to this machine anytime soon, so I hope this helps some of you guys who are as much of a spaz as I am in trying to find the dream machine. lol

Take care,

Pixel Pete

[jrod7350]

Man, according to this page, the G5 is an ABSOLUTE DOG!!!!
What the heck is happening here? So many people who saw pre-release versions of the machines thought they were so fast, but gosh, this is horrible. An Athlon 2200 MP (with a single proccessor) does the gaussian blur in 3.4 seconds compared to 10 seconds for the G5!!!!! That's practically three times slower. And this was the optimized version of PS?!! Ohh, Something is definetly, wrong . . . wrong, wrong, wrong.

This is actually kinda depressing after watching the WWDC bakeoffs. Can't we even get close to that performance?

P.S. BTW, I know the caveat on the RAM amount, but really, could it warp the scores that much?

[moreno]

http://www.wolan.net/~moreno/lindo.jpg

got G5?

[Eug Wanker]

I know the caveat on the RAM amount, but really, could it warp the scores that much?

AFAIK, yes.

[jrod7350]

Originally posted by moreno:
http://www.wolan.net/~moreno/lindo.jpg

got G5?

I'm not sure which G4 that is compared to which P4, but There were cases that I could get 50% more performance in Photoshop out of my G4 than a comprable P4 (back in the 2Ghz P4 days and dual 800 or 1Ghz G4). But c'mon, this is absolutely pathetic. I mean my Dual 1Ghz machine runs faster than this 1.6 G5. There's got to be something wrong here. Flat out.

[Hydra]

If the test ended up using a scratch disk in any way you start testing the hard drive instead of the cpu/memory systems. To quote the page you refer to:

"We found out that our benchmarking became much more repeatable and reliable when we increased memory usage to "81%", which was exactly 375 MB. This way no harddisk swapping could occur."

This is more memory alloted to PS than that G5 has total. OSX is arguably a RAM hog to begin with and I doubt people will spend $600 on a Pro app like PS and put it on a machine with a tiny 256MB of RAM.

-Jerry C.

[Brazuca]

With only 256 ram, this test only serves to keep us a bit distracted until a serious tests comes up.

Soon some of these early owners, after receiving and installing their Ram sticks, will start posting their results.

Take a breath.

[freaktornado]

What the heck is happening here?

Um, a touch of hysteria on your part?

RAM has a huge effect on PS, and IMO 256 MB is barely enough to run OS X efficiently let alone OS X and Photoshop.

[businezguy]

Yeah the problem here is that people are going to stores and testing the machines. The machines have only 256 megs of Ram. You really need to have 1 gig of Ram installed on these machines to run these tests. Without it, the process goes to the hard drive and that seriously slows the G5 down. We really need to wait for these things to ship in quantity until we can see the results we are looking for.

[blackwind]

Unfortunately, these G5 results are basically useless (unless someone wanted to know specifically how fast their computer is when Photoshop hits their scratch disk).

The general rule of thumb is to have at least 5 times the file size of RAM free for Photoshop.

Given a 50-MB file, Photoshop must have at least 250 MB of free RAM to avoid the scratch disk. Even 250 MB of free RAM is probably not enough.

If you could test on a machine with 1024 MB of RAM, that would be great.

In any case, thank you for going to all the trouble to run this test for us.

[jrod7350]

Oh well, I'm sure you guys are right, but I decided I'd run PSBench7 on my Dual 1Ghz machine, and yah, it's on average about twice as fast as the 1.6 G5, and that's too bad, cause my Dualie gets eaten by the faster P4's and dual AMD systems.
I think we should stop benchmarking, it's getting too depressing.

[sleepyrenderer]

PS7Bench 50meg ver needs at least 512MB of RAM so it doesn't become mostly a HDD benchmark.

I'm guessing the dual 2.0 will end up being very competitive w/ top of the line dual Xeon systems with respect to Photoshop, which would be quite a decent improvement. As for the WWDC stevenote bakeoffs, we should all know by now those tests are all handpicked best case scenarios and can't be used to infer overall realworld performance.

If you'll remember back in the day Steve had a bakeoff which involved some filters and functions being used on the Monster's Inc. poster. As with every Macworld bakeoff the P4 of course got horribly creamed by the G4 but in third party testing the two systems were actually pretty evenly matched as for photoshop work.

But again, as others have said, these Photoshop numbers aren't useful. Hopefully we'll see some 512MB+ ram scores soon.

[Hydra]

Originally posted by jrod7350:
Oh well, I'm sure you guys are right, but I decided I'd run PSBench7 on my Dual 1Ghz machine, and yah, it's on average about twice as fast as the 1.6 G5, and that's too bad, cause my Dualie gets eaten by the faster P4's and dual AMD systems.
I think we should stop benchmarking, it's getting too depressing.

That's horrible... Maybe you should add some more RAM to your Dual 1GHz because it might bench PS even better. If your dual has more than 256MB RAM then your test is invalid as well because your machine is not running the same configuration as the G5.

Funny I thought you said that that PC's on that link you provided were 3x's faster than the G5. Your G4 is twice as fast as the G5 tested so your mac is competitive with some fast PC's. Why are you depressed? You don't need a G5 at all. Your G4 system is apparently very fast according to all theses developments.

Just goes to show you how benchmarks are of limited use and improperly done benchmarks are virtually useless.

-Jerry C.

[PixelPete17]

Yeah the test was kinda futile, but I was in front of the machine and had no choice but to use what I had. A one gig RAM setup would tell a whole nuther story I'm sure. I ran a few tests on a 12 MB file too, and most things were instantaneous.

For the record, I was also purging the RAM after every filter or command, so as much resources were available to PS as possible.

My guess is with sufficient RAM, the machine will be nice, but not a Mike Tyson knockout puncher of a 'puter. The dualie will be a beautiful machine, but for $3000 grand and after two years of mediocre performance history, it better be.

Looks like we'll just have to keep waiting.... (sigh)

Pixel Pete

[jrod7350]

Originally posted by Hydra:
That's horrible... Maybe you should add some more RAM to your Dual 1GHz because it might bench PS even better. If your dual has more than 256MB RAM then your test is invalid as well because your machine is not running the same configuration as the G5.

Funny I thought you said that that PC's on that link you provided were 3x's faster than the G5. Your G4 is twice as fast as the G5 tested so your mac is competitive with some fast PC's. Why are you depressed? You don't need a G5 at all. Your G4 system is apparently very fast according to all theses developments.

Just goes to show you how benchmarks are of limited use and improperly done benchmarks are virtually useless.

-Jerry C.

Nah, my Dual has 1.5GB of RAM, it's totally maxed. But yah, if these benchmarks are anything to go by, the old, slow G4's are really the best buy.

[Eug Wanker]

Originally posted by jrod7350:
Nah, my Dual has 1.5GB of RAM, it's totally maxed. But yah, if these benchmarks are anything to go by, the old, slow G4's are really the best buy.

Err.. no. What it tells us is that 1.5 GB with a dual G4 is better than 0.25 GB with a single G5. But we already knew that, because we already knew that 256 MB is insufficient to run both OS X and Photoshop with a large file.

[RooneyX]

Could you please reduce the RAM in your Dual 1Ghz to 256 and then run the tests? Then at least we can still make a comparison even if scratch disk space is being used.

[Xaaron Swiftblade]

Hmm, I wonder how much ram Mr. Jobs' dual 2ghz had at WWDC. I betcha there's an '8' somewhere in that number

[Eug Wanker]

Originally posted by RooneyX:
Could you please reduce the RAM in your Dual 1Ghz to 256 and then run the tests? Then at least we can still make a comparison even if scratch disk space is being used.

Well, not really, unless he manages to get the same 25 MB file too, and gets the same hard drive.

[jrod7350]

Originally posted by Xaaron Swiftblade:
Hmm, I wonder how much ram Mr. Jobs' dual 2ghz had at WWDC. I betcha there's an '8' somewhere in that number

No, supposedly the P4 had more RAM than the G5. I don't remember how much RAM was in those WWDC G5's but I don't believe it was over 2GB.

I have a 2Ghz Dual on order, but I must admit, these preliminary benchmarks are really getting me worried. It was a bit of a stretch, and I want to know I'm getting what I was expecting, not some smoke-and-mirrors parlay with a bunch of people faking benchmarks on the platform
While people can be suspect all they want of Steve Jobs, I really don't think that all those other people from the companies would put their reputations on the line by lying. Even Brad Peebler from Luxology admitted that the G5 was a screemer and backed up his test results with a public statement on his website for all the naysayers. So I don't think what we were seeing was wrong, I just want to know where all the performance advantages have gone. I mean that 2Ghz G5 was over twice as fast as the dual Xeon. TWICE as FAST! Right now we're getting three times slower in some cases-even though it's not the best benchmarking conditions, it's still slow. Okay, end of rant.

[moox12]

[QUOTE]Originally posted by PixelPete17:
[B]OK guys,

here's some numbers from test that you will have to replicate on your own G4s on your own time in order to create a table of G4 vs. G5 data.

The machine only had 256 RAM, it DID have the G5 optimizer installed on it (as I was standing right there when I had the sales guy do it).

I only have time to post some of the numbers I have, but overall, I'd say I can't really form a concrete opinion on this machine until I see it running with a 768MB or more RAM. The 256 was fine for everyday tasks, OS X felt crisp, and I'm sure Panther will be a beauty of a smooth OS once it's out.

Here's the numbers, if you hae questions just post em, and I'll do what I can to answer by tomorrow. I'm going out tonight (after dinner), so I'll get back to you when I can.

Here ya go:

25.2 MB file (digital photograph if you care) PS 7.01 with G5 update:

You used an 25.2 MB file and this is very importand. What kind of document is it, a Photoshop, JPG or tiff. And is this 25.2 MB a Finder reading or a PhotoShop file size reading (bottom schreen)
Most files are stored compressed and decrompressed when open and that differnces from the kind of file it is. Please tell us. 25.2 MB isn't always 25.2 MB. Thanks

[daydreamer]

Run the tests with equal memory damnit. OS X alone needs a minimum of 256mb Ram, or just lower the memory in the dual G4 to 256mb Ram..

[Eriamjh]

Originally posted by 1stunna:
hey pete. (i was at the uofa).

i dont think that the 1.6 need to have ram put in in pairs. isnt there only one bank?

I think all G5s require memory in pairs.

The single G5s have 4 memory slots, 2 in each bank.

Since the dual G5s have dual busses, I don't understand how a single G5 can need memory in pairs, but that is the case regardless. Just try to add RAM to a single G5 at the Apple store and you'll see what I mean.

[Bill Harrison]

Man oh man... Just how long is it going to take you guys to realize that the G5 is NOT the be all end all. I mean, its showing to be slower in EVERY test. First its "well this benchmark is not optomized for the G5". Then XBENCH was optimized, and then the excuse became "well xbench is just junk anyways, it will kick ass in real world PS tests!". Now, these are done, and MORE excuses are being made! "It needs more ram, thats what is holding it back!"

Come on guys! If you have to make THIS many excuses for something, dont you think there is a problem?

Now, for my opinion. The G5 IS a great step in the right direction. It has good future possibilities, and the architecture upgrades are great as well. SATA, much faster bus speeds. Love the look of the machine. BUT, Someone over at apple was being VERY creative when they came up with "The Worlds fastest PC". That is what upsets me the most. Apple could have been proud, and said, here we are, with a competitive platform, that has great potential. Instead they had to blow things out of proportion. There were threads about how the DUAL 2ghz was SIMPLY OVERKILL for anything other than supercomputer tasks! how browsing the web on a G5 was going to be a HUGE waste of its potential!

Give me a break. Both intel and AMD know what they are doing, and it appears IBM does as well. But, if apple wants to go into the bar and yell at the top of its lungs that it is the strongest man there, it better be able to back that up! And Apple came in, and now all i see are excuses. Lets just enjoy the G5 for what it is, the first step in the right direction for apple processor wise. I just feel its REALLY poor taste that apple has mislead people on this issue, although it is NOT the first time.

[Eugene]

Originally posted by Eriamjh:
I think all G5s require memory in pairs.

The single G5s have 4 memory slots, 2 in each bank.

Since the dual G5s have dual busses, I don't understand how a single G5 can need memory in pairs, but that is the case regardless. Just try to add RAM to a single G5 at the Apple store and you'll see what I mean.

I don't understand what you mean. The usefulness of dual-channel memory configurations has NOTHING to do with how many CPUs a computer has. On the 1.8 GHz G5 for example. there's a 900 MHz link from system controller to CPU. That's 6.4 GBps (7.2 theoretical) of bandwidth available to the CPU. ...3.2 up, 3.2 down actually...

PC3200 memory in a single-channel configuration provides up to 3.2 GBps of bandwidth. In a dual-channel configuration bandwidth to the memory is effectively doubled.

Also, in the dual G5s, it's not a shared bus topology so each CPU gets over 6.4 GBps of bandwidth.

Or am I wrong?

[Eug]

Since the dual G5s have dual busses, I don't understand how a single G5 can need memory in pairs, but that is the case regardless.

All G5s have dual channel memory busses. The busses connecting the CPUs are independent from the memory busses.

Then XBENCH was optimized, and then the excuse became "well xbench is just junk

According to the Ars programmer type, that's true. They used to say that even in the G4 days.

"It needs more ram, thats what is holding it back!"

And this is true as well. Before the tests were even run, people were saying 768+ RAM was needed to make the test meaningful. Tests have been run on G4s before where some machines had insufficient RAM. Those machines had times FOUR TIMES as long (or even longer). And it makes perfect sense that it was much slower, because paging out to disk eats up time. If it didn't, there would be no point for extra RAM in a computer.

how browsing the web on a G5 was going to be a HUGE waste of its potential!

And again, this is true. Just browsing on the web on a G5 is a HUGE waste of its potential, and indeed, huge waste of money. Buy an eMac for that.

[xMetal]

I honestly think that people get so involved in "benchmarks" and crap like that, that they completely lose perspective on the overall usability of a machine.

I have a 17" Powerbook. It has a "lowly" G4, and only one of them. I upped the ram to 1Gb, and this combination makes for a plenty fast machine. I am a professional designer/content creation guy, and run serious software all day long, to do tasks as varied as 3D modeling, video editing, web design, SERIOUS photoshop work, page layout and more. Often all at the same time.

I rely on this machine as my sole source of income, basically, and it works JUST FINE. Sure, I might have to wait a second or so longer doing something, but it's really not that bad.

The G5, even in it's slowest incarnation, will completely decimate my Powerbook, yet you guys are freaking out. Honestly, how many of you, being COMPLETELY HONEST, need that speed. Don't BS about it. If you do a little design work, surf the web, do your email, etc and are flipping because one test of a G5 shows it doesn't have Uber-power-from-heaven, then you really need to evaluate your life. Relax. You'd be surprised what real people who use their computers for a living get by with.

And if you really do need the power, you'll get a Dualie with a couple gigs of ram anyways, because the machine will be your livelyhood.

Geez people. No need to have an anuerysm over a computer. Deep breaths my friends.

[Anand]

"This build, code-named Smeagol, will run on the new chip but won't take advantage of many of its key features, including 64-bit support. Sources said Apple's goal for Smeagol is to deliver Mac OS X performance at least "on par" with what Jaguar could achieve on Motorola G4 chips running at the same speed;"

From this article:

http://www.eweek.com/article2/0,3959,1121736,00.asp

It is exactly what eweek said months ago. The G5 will no shine untill Panther. Until then, it is just a faster G4.

[jrod7350]

Originally posted by Bill Harrison:
Apple could have been proud, and said, here we are, with a competitive platform, that has great potential.

Uh, as of right now, the G5 isn't even competitive with older G4's, forget the wintel crowd. Potential, maybe, but as of right now, the performance is absolutely abysmal. (oh wait, I forgot, there wasn't enough RAM)

[lebannen]

No, no, no, no, no.

PixelPete17 - thanks for running the tests. However, these tests are more like running a set of arbitary tasks - quoting the numbers you get is completely misleading and seems to have depressed a few people

To everyone: You CANNOT compare those posted figures to the AMD tests, or to tests run on your home G4, unless you are doing exactly the same operations on each machine. This is why the Ace's Hardware tests use PSBench - this uses a standard file and performs recorded operations on that file, so that parameters are kept as exact as possible.

Fact: You can't compare a 10-second gaussian blur of a 23 Meg file to a 0.2-escond gaussian blur of a 50 Meg file. The 23 Meg be actually be a jpeg, uncompressing to several hundred megs; the dimensions are different; the contents of the picture are different; and the parameters of the picture are different.

So, PixelPete, the community is crying out for some real benchmarks Would you mind giving PSBench a run on that machine? (For the uninitiated, PSBench is a set of standard actions on specific files, and includes benchmark parameters such as screen res. This is what sites like ace's hardware use, because then you can compare different machines more accurately). Otherwise, how does Photoshop 'feel' on that machine?

But until we see a standard test run, the results above are pretty useless for comparing with any other machine

[SomeToast]

Originally posted by Bill Harrison:
how browsing the web on a G5 was going to be a HUGE waste of its potential!

It is. Browsing the web on my dual-GHz G4 is a huge waste of its potential.

What's with all the new members whose sole purpose seems to be to say how concerned they are over the G5's "abysmal" performance?

[Eug]

I have a 17" Powerbook. It has a "lowly" G4, and only one of them. I upped the ram to 1Gb, and this combination makes for a plenty fast machine. I am a professional designer/content creation guy, and run serious software all day long, to do tasks as varied as 3D modeling, video editing, web design, SERIOUS photoshop work, page layout and more. Often all at the same time.

I rely on this machine as my sole source of income, basically, and it works JUST FINE. Sure, I might have to wait a second or so longer doing something, but it's really not that bad.

That's interesting, because I'm NOT a professional designer/content creation guy. I do stuff with images for work, but my photoshop needs are light. I dabble with video, just with iMovie and iDVD... And I find the 1 GHz G4 just medicore for still image work, and slow for video.

I will upgrade hopefully by mid to late 2004. However, if a 1.8 G5 PowerBook existed today (with good battery life), I'd upgrade tomorrow. The 1 GHz G4 is merely OK for what I do as an amateur. I can understand how others would want something MUCH faster. (I could get a faster PC laptop, but I don't like the OS, and overall I'm more productive on an OS X Mac. However, it'd be even better if it were even faster.)

I'm sure glad I don't have to do 3D modeling on this machine. I'd probably rip my hair out.

[I Me Mine]

The thing that concerns me about benchmarks is this:

At WWDC everyone was delighted because the benchmarks showed the G5 was beating Xeons and Pentium 4s.

Now, because some early results so far have been lower than people thought, benchmarks are "worthless".

Plus, this whole business with Xbench scores being low leads people to say "Well, Xbench is crap anyway" - What would have happened if the scores for 1.6 G5 were incredibly high? Would Xbench be crap then?

I think the real performance of the G5 is somewhere between WWDC in June and Xbench today.

[Eug]

What would have happened if the scores for 1.6 G5 were incredibly high? Would Xbench be crap then?

Yes, according to the Ars people. They had said so long before the G5 even came out.

[xMetal]

That's interesting, because I'm NOT a professional designer/content creation guy. I do stuff with images for work, but my photoshop needs are light. I dabble with video, just with iMovie and iDVD... And I find the 1 GHz G4 just medicore for still image work, and slow for video.

I guess it's all in your perception, and your personality.

I don't plan on having an ulcer by the time I'm 30, so waiting an extra few seconds doesn't really get to me. I can also remember doing 3D on a Mac IIci with an unbelieveable 12mb of ram, so what I can do now is pretty darn nice.

I"m sure I'll upgrade too. My point is that far too many people absolutely FREAK about how fast/slow a machine is, without even thinking about their usage habits. There's far too much "I must be faster than the other guy to prove I'm a worthwhile human being" kind of mentality these days, in nearly everything. Witness rising car horsepower levels for more evidence of that. People that don't know how to handle a 70hp Geo Metro just have to have that 300hp sedan...

[I Me Mine]

Eug - You must admit though, there wouldn't be as much doom and gloom around as there are with some people.

[daydreamer]

Heh, so now we need tests of a G5 with the latest Panther build and min 512mb RAM compared to a dual G4 machine with equal RAM... I don't think numbers will change that much.. but if 10.2.7 really has zero 64bit support then numbers should a lot higher right??

[Hi I'm Ben]

Why don't folks wait until at least the panther release till you start freaking out. I haven't even seen any solid benchmarks and evidence yet.

OMG!GJNHJKH!J@HJ!NOOOOOOOOOOOOO THE G5 SUx0R! TELL Y3R FRENS!

[Bill Harrison]

Read carefully. I never said abysmal performance. I said performance in line with current offerings. Which is not slow. I just think its BS that Apple pulled that Fastest PC in the world crap.

Now, that said, i would love a REV 2 G05 With a bit more speed and dual processors. Because OSX is the best thing i have ever used, regardless of speed.

So before you call anyone a newbie with a bad attitude, just realize im pointing out what I feel is crap, not picking on the G5 in general.

[Eug]

Eug - You must admit though, there wouldn't be as much doom and gloom around as there are with some people.

True, true.

I don't plan on having an ulcer by the time I'm 30, so waiting an extra few seconds doesn't really get to me. I can also remember doing 3D on a Mac IIci with an unbelieveable 12mb of ram, so what I can do now is pretty darn nice.

Heh. Perhaps some of it is my usage pattern. My image manipulation is just a side thing for work. So it gets left for the last minute because I don't get paid extra to do it. And at the last minute I grow restless as the seconds tick by...

As for iMovie 3 and iDVD 3, I just find the jumpiness and time lag irritating.

[MindFad]

Originally posted by Eug Wanker:
Well, not really, unless he manages to get the same 25 MB file too, and gets the same hard drive.

Yes, people, calm down. These are scratch disk benchmarks if anything. Put a gig in there and she'll fly.

[Bill Harrison]

Originally posted by xMetal:
I guess it's all in your perception, and your personality.

I don't plan on having an ulcer by the time I'm 30, so waiting an extra few seconds doesn't really get to me. I can also remember doing 3D on a Mac IIci with an unbelieveable 12mb of ram, so what I can do now is pretty darn nice.

I"m sure I'll upgrade too. My point is that far too many people absolutely FREAK about how fast/slow a machine is, without even thinking about their usage habits. There's far too much "I must be faster than the other guy to prove I'm a worthwhile human being" kind of mentality these days, in nearly everything. Witness rising car horsepower levels for more evidence of that. People that don't know how to handle a 70hp Geo Metro just have to have that 300hp sedan...

Geo metros have 47 Horsepower.

Trust me, I know!