New Intel iMacs tested vs G5, Jobs was way off the mark

Mastershakes

OSNN Veteran Addict
#1
MacWorld

Macworld Lab’s tests do show that the new Intel-based iMac is faster than the iMac G5 when running native applications. However, we found that those improvements are generally much less than what Apple claims is a 2x improvement in speed.

Instead, our tests found the new 2.0GHz Core Duo iMac takes rougly 10 to 25 percent less time than the G5 iMac to perform the same native application tasks, albeit with some notable exceptions. (If you'd prefer, that makes the Core Duo iMac 1.1 to 1.3 times as fast.) And we also found that applications that aren’t yet Intel-native—which must run using Apple’s Rosetta code-translation technology—tend to run half as fast as the same applications running natively on the iMac G5.
Makes sense. Too bad, Apple has a fairly pristine image in my books, it sucks they lied to us.
 

sean.ferguson

OSNN Veteran Addict
#2
sure common tasks such as rendering and opening documents in say Pages or Keynote will not be 2x faster. The precise applications that Apple refer to are Integer and Floating Point tests, so in theory the numbers that Apple provided may very well be correct.

In every day use the MacBook and the iMac are not going to perform 5x and 2x faster respectivly.
 

sean.ferguson

OSNN Veteran Addict
#5
Mastershakes said:
NOTHING runs at 2x on it.

I'd say that's an incorrect.
Throw in an extra 512MB RAM into those test machines and they probably will. Tests are always variable, tests such as these can never be conclusive.
 

Mastershakes

OSNN Veteran Addict
#7
sean.ferguson said:
Throw in an extra 512MB RAM into those test machines and they probably will. Tests are always variable, tests such as these can never be conclusive.

I'll believe it when I see it. Being Macworld I would imagine they would be relatively unbiased.

RAM won't make a difference. It's the programming, the emulation that is the issue.
 

X-Istence

*
Political User
#8
Mastershakes said:
Makes sense. Too bad, Apple has a fairly pristine image in my books, it sucks they lied to us.
Uhm, Apple has not lied to you. They specifically tell you in what regards they are 2x faster. They were not talking about the speed of applications and whatnot, but how fast it could perform the SPEC test:

Testing conducted by Apple in December 2005 using preproduction 20-inch iMac units with 2GHz Intel Core Duo; all other systems were shipping units. All scores are estimated. SPEC is a registered trademark of the Standard Performance Evaluation Corporation (SPEC); see www.spec.org for more information. Benchmarks were compiled using the IBM compiler and a beta version of the Intel compiler for Mac OS.
(http://www.apple.com/imac/, footnote number 1)

And here is the chart they used to base their information on:
Code:
iMac G5 2.1 Ghz:

     SPECint_rate_base2000 (Integer calculation(estimate)):
          10.2
     SPECfp_rate_base2000 (Floating-point calculation (estimate)):
          13.0

iMac Core Duo 2.0GHz:

     SPECint_rate_base2000 (Integer calculation(estimate)):
          32.6
     SPECfp_rate_base2000 (Floating-point calculation (estimate)):
          27.1

∆:

     SPECint_rate_base2000 (Integer calculation(estimate)):
          3.2
     SPECfp_rate_base2000 (Floating-point calculation (estimate)):
          2.1
There is no false advertising, and no lying here. It would also be very bad if the dual core Intel did not outperform a single G5 in any of these operations. However, day to day applications don't do just integer or floating point calculations, that is why a 2 - 3x speedup won't be seen in day to day applications, when it comes to pure processing power with regards to calculations, the Intel mac's beat the G5's hands down. Another reason the speedup won't be seen is in fact because the Intel and G5 ways of accessing memory are almost the same, it is not like AMD where the memory controller is on the chip, but rather the memory controller is on the north bridge, which introduces some latency. (Don't quote me on this, this is just from memory, and I could well be wrong). Not only that, but most apps will not be taking the full advantage of the dual core goodness that Apple is providing to them, as most of them still run in a single thread. The only way an Application can make use of the two cores is to be multi-threaded, and that is most of the time more work, than it is worth for software developers. I sincerely expect the two iMac's to be on par with each other. I never expected it to be faster in starting applications or day to day applications.

Where we will see the Intel Mac's slow down is when it comes to doing graphical work, even with universal binaries. SSE3 is just not up to par with AltiVec when it comes to processing data in a vectorized manner. So professionals that are working on high end video projects will for a while to come, see more power in G5's while doing work than an Intel Mac. Apple knows, this, and that is also the reason why the top of the line PowerMac's have not been replaced yet.

I hope this explains just a bit how Apple did the tests and where the numbers are coming from. If Steve Jobs said in his keynote "Applications launch at the same speed as before, and there are no slowdowns" it would not have been as great of an achievement. Welcome to marketing.
 

X-Istence

*
Political User
#9
sean.ferguson said:
Throw in an extra 512MB RAM into those test machines and they probably will. Tests are always variable, tests such as these can never be conclusive.
Nah, they then even won't run 2x faster. Integer and floating point tests are a bad way to market how fast a machine is, but that is marketing. Show good big numbers that make people jump for joy and add a small footnote somewhere. The numbers are accurate, just not for what people really use the machine for.
 

LeeJend

OSNN Veteran Addict
#10
HEY, it's called marketing, not lying.

Just because they used a benchmark test designed to rate the scientific algorithm processing capability of mainframes to sell their general purpose, business application, laptop product does not mean they are trying to mislead their customers...

Well, wait, actually it does.
 
#11
Mastershakes said:
Too bad, Apple has a fairly pristine image in my books, it sucks they lied to us.
Pristine image? I hope that was sarcasm, because this is definitely not the first time they've done something like this, and it most certainly won't be the last. See the link to an independent test conducted by AnandTech towards the end of this thread. Compare the results to Apple's published benchmarks.

Heard of the Reality Distortion Field? ;)
 

X-Istence

*
Political User
#12
NetRyder said:
Pristine image? I hope that was sarcasm, because this is definitely not the first time they've done something like this, and it most certainly won't be the last. See the link to an independent test conducted by AnandTech towards the end of this thread. Compare the results to Apple's published benchmarks.

Heard of the Reality Distortion Field? ;)

It has sort of been fused into what Apple is about. Most people know not to drink the Kool-Aid, and step outside of the reality distortion field. The G5 tests were flawed and biased in a lot of ways. On the G5 at the time they used a special IBM compiler, on the Intel machines they used GCC, and not the Intel compiler. On the G5 they also optimized the code created to use a smaller amount of data in memory or something or other, while this was not done on Intel systems.
 
#13
NetRyder said:
Pristine image? I hope that was sarcasm, because this is definitely not the first time they've done something like this, and it most certainly won't be the last. See the link to an independent test conducted by AnandTech towards the end of this thread. Compare the results to Apple's published benchmarks.

Heard of the Reality Distortion Field? ;)
Lets not forget that every CPU maker (as well as graphics cards) have been caught doing the same exact thing. Only some them really did lie instead of just kinda bending a statement to fit what everyone wants to hear.
 

Mastershakes

OSNN Veteran Addict
#17
NetRyder said:
Pristine image? I hope that was sarcasm, because this is definitely not the first time they've done something like this, and it most certainly won't be the last. See the link to an independent test conducted by AnandTech towards the end of this thread. Compare the results to Apple's published benchmarks.

Heard of the Reality Distortion Field? ;)
I should start paying more attention. My bad. Will check out links Net, thanks.
 

sean.ferguson

OSNN Veteran Addict
#19
the whole G5 thing was stupid - they made the G5 sound like an amazing chip that was going to destroy the world, where in actual fact it just wasn't as big as they made out.

Now I have used a G5 iMac, and I have to say it isnt all that damn amazing. It felt slower than my Sempron machine in doing the usual day-to-day tasks.
 

Sazar

F@H - Is it in you?
Staff member
Political User
#20
sean.ferguson said:
the whole G5 thing was stupid - they made the G5 sound like an amazing chip that was going to destroy the world, where in actual fact it just wasn't as big as they made out.

Now I have used a G5 iMac, and I have to say it isnt all that damn amazing. It felt slower than my Sempron machine in doing the usual day-to-day tasks.
Yep, I said this almost a year or so ago when comparing things to my a64 3400+ at the time.
 

Members online

No members online now.

Latest posts

Latest profile posts

Perris Calderon wrote on Electronic Punk's profile.
Ep, glad to see you come back and tidy up...did want to ask a one day favor, I want to enhance my resume , was hoping you could make me administrator for a day, if so, take me right off since I won't be here to do anything, and don't know the slightest about the board, but it would be nice putting "served administrator osnn", if can do, THANKS

Been running around Quora lately, luv it there https://tinyurl.com/ycpxl
Electronic Punk wrote on Perris Calderon's profile.
All good still mate?
Hello, is there anybody in there? Just nod if you can hear me ...
Xie
What a long strange trip it's been. =)

Forum statistics

Threads
61,971
Messages
673,299
Members
89,016
Latest member
Poseeut