Comparative G5/G4 Tests 102
rocketjam writes "Barefeats.com has posted test results comparing a 2GHz G5 MP, 1.8GHz G5, 1.6GHz G5 and G4 MP's at 1.4GHz, 1.25GHz and 1GHz. They use Photoshop, Cinebench 2003 and a Bryce 5 render for tests. Bottom line is the G5 2GHz MP has the best bang for your buck, but you might think twice about trading in that dual processor G4 for a solo G5...the G4s hold their own quite well. They also say tests in Panther yielded significant increases in the G5 scores."
Hmm. Not much of a review. (Score:2, Interesting)
We ran Xbench 1.1 on a G5 1.8GHz with 10.3 beta build 7B49. Compared to 10.2.7 "Jaguar"....
Hmm. I guess that Apple users aren't too discriminating when it comes to stats or methodological testing descriptions in a review! (Perhaps, though, with Anandtech and Tom's Hardware, we're jus
Re:Hmm. Not much of a review. (Score:5, Insightful)
In general though, testing procedures for Macs is nothing like it is for PC's. Too bad Anandtech and others wouldn't put some focus on Mac's from time to time. *shrug*
Re:Hmm. Not much of a review. (Score:4, Insightful)
The G5 is NOT going to excel at Altivec optimised code, the G4 will remain the Altivec champ until IBM puts the kind of Altivec resources onto the 970 that Moto put onto the 745x series. For most other stuff, the G5 will wipe the floor with the G3 and G4 chips.
Re:Hmm. Not much of a review. (Score:5, Informative)
That's not strictly true, the problem is that altivec that's optimal for the 745x chips is worst case for the 970, since best case on the 745x makes heavy use of the vec_dst (prefetch) instruction, which is handled serially on the 970 and hence stalls the Altivec hardware.
also the G5 is a lot less likely to run into bandwidth bottlenecks with altivec code on large datasets.
Re:Hmm. Not much of a review. (Score:2, Informative)
Re:Hmm. Not much of a review. (Score:3, Informative)
Re:Hmm. Not much of a review. (Score:3, Informative)
If you read Hannibal's appendium to his 970 articles he points out that the Altivec looks better than he initially thought. This isn't to say there isn't room for improvement. But to say that the G4 is the Altivec champ seems somewhat incorrect.
Hannibal's Followup [arstechnica.com]
It is true that with G4 altive
Re:Hmm. Not much of a review. (Score:5, Informative)
Another poster has already mentioned the execution sterilization that occurs thanks to optimizations that worked quite well in the G4 (vec_dst et al), but there is another point on this issue that needs to be mentioned.
Even if the G5 performs slightly slower than the G4e with AltiVec enhanced code on an individual, it has the potential to be much much faster overall simply due to the architecture. One of the main problems with AltiVec on the G4e currently is that many of the apps that take advantage of it are
re: hmmm. (Score:5, Funny)
Did you just imply that Anandtech and Tom's hardware are valid places to go to learn about PC hardware? LMAO. For the uninitiated, here's a quick rundown of every Tom's hardware and anandtech page.
Motherboard X just came in. It's Company X's newest board, poised to grab the speed crown from Company's Y. Can it do it? Read on! (click banner ad.)
Next Page: Here is a blue line that represents motherboard X. You can see that it barely outdistances Y in this key test, as shown by the shorter line. But what about this other test? (click through for more huge banner ads.)
(click through for more huge banner ads.)
(click through for more huge banner ads.)
(click through for more huge banner ads.)
(click through for more huge banner ads.)
Conclusion: Boy, Motherboard X sure is fast! I can't wait to get my hands on one again, and I'll surely pay a premium for such speed. There may be banner ad for a company selling this motherboard RIGHT ON THIS VERY PAGE! I know what I'd do, if I wanted to be l33t, and you want to be l33t, don't you?
(click through for more huge banner ads.)
Next week: Will the red line finally beat the blue line in our extensive test suite?
Not hard to do (Score:4, Interesting)
Re:Not hard to do (Score:5, Interesting)
The G4 is cheaper.
The G4 does multiprocessing for the price.
The G4 is able to boot OS 9. The G5 isn't.
The G4 is compatible with my low voltage PCI cards. The G5 isn't.
The G4 is available today. The G5 isn't.
The G4 ist a pleasure to look at. The G5 isn't.
Why on earth should I choose a G5 over a dual G4?
Re:Not hard to do (Score:4, Informative)
The G5 is faster.
The G5 accommodates four times the RAM (roughly), and the RAM is considerably faster.
The G5 has AGP 8x.
The G5 has USB 2.0.
The G5 has FireWire 800. (The last generation G4 did, but the generation currently being sold by Apple is next-to-last and so does not.)
The G5 has Airport Extreme. (Ditto.)
The G5 has Bluetooth. (Ditto.)
The G5 has internal serial ATA.
The G5 has digital audio I/O.
The G5 is considerably quieter.
If none of that stuff matters to you, get the G4. If it matters, get the G5. This isn't a clear-cut case of "G5 better than G4 in every way." But neither is it even remotely a case of "no reason to buy G5."
You forgot one... (Score:3, Funny)
(I suppose you could grate cheese with the fan on the back of the G4, but that would get messy real quick!)
Re:You forgot one... (Score:3, Funny)
And grinding a hunk of cheese against the front of a G5 isn't going to get messy? There's all kinds of fans and shit behind that apparent cheese-grater front, and you know that they'll be picking up bits of half shredded cheese and spraying it all over the RAM and the heat sink and melting all over everything.
Nooo, grating cheese on a G5 would be much messier than grating cheese on a G4.
Re:You forgot one... (Score:1)
Re:You forgot one... (Score:1)
Actually I like the dual function that the G5 provides: grating and melting cheese.
Re:You forgot one... (Score:2)
Re:You forgot one... (Score:1)
I think it looks a lot more like an unrolled surface from a micro-screen electric razor.
Re:You forgot one... (Score:2)
actually, the holes are raised, it's just the metal surround thats flat
dave
PCI-X (one other thing) (Score:4, Informative)
Re:Not hard to do (Score:3, Insightful)
It doesn't run OS 9, however it does run Classic, and from a security standpoint OS 9 is a problem, and most Mac viruses are written for OS 9.
Unless you have a specific application that you must be running in OS 9 and that is unable to run in Classic mode on OS X, or specific legacy hardware that is not supported in OS X (or by the new G5) From a long term viability standpoint the G5 will be a better alternative.
(unless of course you must run VPC 6.1 or earlier on the mac which
Re:Not hard to do (Score:3, Informative)
Re:Not hard to do (Score:2, Insightful)
Though I'd love to have digital audio I/O, for video editing more internal drive bays is winning out. And that I have a sizeable investment in non-serial ATA drives and the video data upon them.
Add to that that I still can't find any external Firewire drive enclosures with large (>137 GB, >128 GiB) drive support (existence so far is only rumored, not substantiated).
Re:Not hard to do (Score:1)
http://www.westerndigital.com/en/products/com
Re:Not hard to do (Score:3, Informative)
Re:Not hard to do (Score:1)
Seems the only way to be sure is to get an enclosure that already contains a large drive, buy from a place that accepts returns, or trust the experiences of strangers. And I'm having to resort to the lattermost.
Re:Not hard to do (Score:2, Interesting)
Re:Not hard to do (Score:2)
Can you please provide your reference for this claim? I find this quite hard to believe. Maybe you meant 'high' and not 'low'?
Re:Not hard to do (Score:2)
"5 V keyed or signalling cards do not work in the Power Mac G5 computer". Basically, many existing PCI cards will not fit the PCI connectors of the Power Macintosh G5. You will have to buy an external PCI chassis if you want to use these cards (and the DC30 series cards) with a G5. According to Apple, this applies to all G5 computers; according to other sources, it applies only to the 1.8 and 2 GHz, while the 1.6 GHz has no limitations in acceptin
Re:Not hard to do (Score:2)
Yeah, it looks like their PCI-X slots are 3.3V only. Not a big surprise. Still, I'm looking at my Big Pile O PCI Cards and very few of them are keyed for 5V only. The ones that are 5V-only aren't the kind of thing that I would put into a $2K machine (do I really want to put a RealTek 10BaseT card in there? Or a Mylex DAC960 UltraWide RAID? Ugh!). YMMV.
Re:Not hard to do (Score:2)
I really want to keep my miroMotion DC30plus frame grabber. That thing once cost a fortune... *sigh*
Re:Not hard to do (Score:2, Insightful)
Re:Not hard to do (Score:2)
Re:Not hard to do (Score:2)
Re:Not hard to do (Score:1, Funny)
bryce render test??? (Score:5, Interesting)
And I find it hard to believe that a 20 second render test will show much insight into a machines speed.
Why not go with a more realistic scene render that would take a couple hours??
Re:bryce render test??? (Score:1)
Photoshop is one example. The code is optimized for the mac. My friends P4 2.8ghz is only a little faster than my G4 733. Apple took advantage of this for years and had test results based on photoshop.
Complex software is not a good way to test different platorms because the code and therefor performance is different.
Re:bryce render test??? (Score:2)
I would imagine the code base for Maya (if it hasn't changed significantly since 1.x) was easier to move to Linux and OS X than the initial Windows port.
Old Myth (Score:5, Informative)
Your working with outdated information--by about 7 years.
The code is optimized for both platforms very heavily. A large portion of Photoshop's userbase is on Wintel machines and they were also paid (by Intel) awhile back to optimize their filters for Intel processors. Presumably they care about performance and keep both versions highly optimized. Saying that it is "more highly" optimized for the Mac is a fallacy--it hasn't even had more time to become optimized for the Mac, due to processor changes. The optimizations for the G3 and G4 are
I also heard a rumor, though I am not sure that it is true, that Photoshop would offload its interface to the graphics card for display on its own and that, with the introduction of QE, Photoshop was effectively rendering its display elements twice. This may be completely unsubstantiated though
While in general your point is good--that apps which are highly optimized for one platform are not necessarily highly optimized for another--Photoshop is a bad example.
Weak article.... (Score:1, Funny)
What about Panther gains for the G4s? (Score:5, Insightful)
Re:What about Panther gains for the G4s? (Score:3, Insightful)
But MacWarehouse, as of last week, had about 20 of them left at under $1,700. My company bought one for me. It's a great machine, and the bang per buck is unbeatable.
I'm getting a G5 for my home, but that's because I do heavy video editing and effects, which should benefit a LOT from the faster system.
But for most things the dual G4's still a great system, especially at that price.
D
Re:What about Panther gains for the G4s? (Score:4, Insightful)
I'd be more excited about the increased memory capacity. Last time I did any clip manipulation on my Powerbook, the swap time was my real killer.
Re:What about Panther gains for the G4s? (Score:3, Informative)
And you're not concerned with the 50% reduction in internal drive bays as compared to the G4? I already have 650 GB shoehorned into my Blue & White G3 for video (upgraded to a G4 processor), and that's not counting the boot drive. The G5 only has two hard drive bays and one optical drive bay compared to the four and two in the G4.
Re:What about Panther gains for the G4s? (Score:3, Interesting)
At the same time, it's either take the speed increases and live with the lower drive count, or don't take the speed increases. And remember, FireWire 800 should finally be fast enough for video editing, especially on a separate bus from the output device.
I could get whatever case I wanted if I switched to the PC world, but then I would have to deal with tiresome Windows problems, and I wouldn't get to use Final Cut Pro, still the best vid
Re:What about Panther gains for the G4s? (Score:1)
Re:What about Panther gains for the G4s? (Score:3, Interesting)
a few rumor sites and mailing lists have reported (hrmmm) that in their personal unofficial tests they thought 10.3 beta was showing the biggest speed increases in CRT iMacs and the older G4 and G3 machines. granted they did not have G5s to test on, but they loaded their beta versions of 10.3 on everything from zippy G4 towers to laptops to CRT iMacs and said they felt the iMacs speed increase felt the greatest. i am curious to try it in my G4 tower (400mghz with a recentl
caveats about these tests (Score:5, Informative)
A lot of programs are yet to be at all tweaked for the G5s. There are a couple of new processor functions (hardware square root for instance) that may bring big gains. Also, there are old functions from the G4 that slow down the G5. Check out Apple's G5 performance primer [apple.com]. I read that the changes are enough to make the guy who made the mac rc5 engine want to re-write it for the G5 (no simple tweaks). He was hoping to get by with just some minor tinkering, but the chip does require a lot more than that to take advantage of it's potential.
Let me take another example; regarding CineBench.
http://www.postforum.com/forums/read.php?f=6&i=87
In that post, a guy called Richard from the Maxon development labs says this:
This compares to a 1.8ghz G5 score that I've seen of 188. Which means they're aiming to get a fair bit out of optimisation for the chip. Just as a means of further comparison...
http://www.imashination.com/bench.
You'll see the top score these people have recorded is for a dual Xeon 2.4ghz - with a score of 502. If the G5s make it up to over 500, that says a lot about the chip.
Finally, you've also got to include a mention of the compilers. Whilst some optimisation has been made to GCC, the GCC guys rejected a whole heap of improvements for the G5 because they were too platform specific. There's a good thread over at Ars Technica [infopop.net] that discusses some pretty big gains when using the IBM XLC compiler. Other Mac-specific compilers should yield some pretty awesome gains too.
So, in summary - take these scores with a grain of salt. They're just the beginning.
-- james
Re:caveats about these tests (Score:2, Informative)
Okay, I'm not going to be lazy. I went to Dell.com and configured a 2.4ghz Dual Xeon with Windows XP, 512mb RAM, a DVD drive and a modem. This is roughly the same configuration offered by the G5.
Price is $2,801.
So the Mac is about $200 more than a system with about the same performance once the Mac is optimized.
It's not going to convince me to use Windows (ugh!), but it's not as good as we were hoping either.
D
Re:caveats about these tests (Score:2)
BTW, Tiger Direct [tigerdirect.com] has Dual 2.4ghz Xeons for 1600 bux.
Re:caveats about these tests (Score:4, Insightful)
Re:caveats about these tests (Score:5, Interesting)
OK, so I am even less lazy, and I disbelieve. I did what you said you did, and my dual-Xeon Precision 650 with a USB mouse, USB keyboard, dual-monitor DVI capable video card 120 GB IDE drive (largest IDE drive there; Serial ATA is not an option and the SCSI drive prices gave me the giggles) and DVD-RW/CD-ROM 48X (hope it's writable) drive and no Dell support (to equal the Apple G5) would cost me a total of $3054.10.
Now, I was too lazy to find out if the Dell had gigE ethernet like the Mac, but this is a closer comparison *IF* you are right about the fact that an optimized dual 2 GHz G5 is only equivalent to a 2.4 GHz dual Xeon. If it actually turns out to be anywhere close to a 3.0 GHz dual Xeon, we can stop having this conversation right now, since that box starts at over $4000. Myself, I'm guessing the truth is somewhere in between.
So that makes the G5 about the same performance as a Dell that might run anwhere between $0 and $500 more. "Priced very attractively" is how I would have to put this. Likely not attractively enough to get *many* Windows shops to switch (based on hardware cost and software legacy alone anyway), but attractive enough to keep market share and probably then some.
Re:caveats about these tests (Score:3, Informative)
Great, except Apple's real-world app tests show the dual G5 spanking Dell's dual 3.06GHz Xeon rig, which when comparably configured (at the time, though it probably hasn't gotten much cheaper since) cost $3772. [slashdot.org] So how is your dual 2.4GHz Xeon going to be "about the same performance" when the faster dual 3.06 Xeo
Re:caveats about these tests (Score:2)
We'll just have to wait and see for something more real-worldish.
D
Comparing Apples and... (Score:3, Interesting)
It looks like the Dual G5 2GHz has the best bang for the buck.
Yeah, it's a big bang but it's also a heck of a lot of bucks and I'm not yet convinced it's worth it. Of course, if (or more likely, when) Apple discontinues the G4-based Powermacs, we won't have a choice.
Re:Comparing Apples and... (Score:1)
For you, maybe its not. But for anyone to whom time is money, that would be a silly question.
Re:Comparing Apples and... (Score:2, Informative)
Re:G5 upgrade woes (Score:2, Funny)
Re:G5 upgrade woes (Score:1)
Re:G5 upgrade woes (Score:2)
"We" can hope? Be very careful what you wish for, and please stop including the rest of us in those dreams!
Re:G5 upgrade woes (Score:1, Redundant)
This is nothing new... (Score:4, Insightful)
real world benchmark (Score:3, Funny)
Re:real world benchmark (Score:3, Funny)
Sure, the G5 is cool... (Score:5, Funny)
Re:Sure, the G5 is cool... (Score:1)
Re:Sure, the G5 is cool... (Score:2, Funny)
Re:Sure, the G5 is cool... (Score:2)
I thought the blue guys had something to with wearing gay butterfly outfits and invading New York. Wasn't that some new scenario proposed by the Dept. of Homeland Defense?
-T
I think the tests bode quite well... (Score:4, Insightful)
Pile that all on top of the speed increases seen under Mac OS 10.3 and you'll get a box that absolutely screams for Photoshop work (let alone anything else). With the tests performed today at bare feats, we can see that the G5 can beat the G4 at it's own game, let's see what the G5's game looks like.
non-optimised software (Score:5, Interesting)
The photoshop tests indicate version 7.0, while Adobe has posted some optimised libraries and plugins for 7.0.1 that supposedly speed up some functions considerably.
I haven't heard back as to whether the author actually used 7.0.[0] or the slightly optimised 7.0.1.
so far, I'm pretty happy with the reported G5 performance considering how different the architecture is to the G4.
Apple's in a very odd position for a Computer Company right now. Instead of releasing increasingly bloated software that runs ever slower, requiring ever faster hardware... their hardware is getting faster and the software is speeding up. i'm running Panther now and it is noticeably faster at many tasks.
Re:non-optimised software (Score:2, Interesting)
Re:non-optimised software (Score:3, Interesting)
Are Benchmarks Premature? (Score:3, Insightful)
Somewhat (Score:3, Informative)
More?! (Score:3, Insightful)
1Ghz FSU, dual 2Ghz 64-bit CPUs, 8GB RAM, serial-ata drives
All that for less than a UltraSPARC server, and you _still_ (!) need "proof" or validation???
This is getting ridiculous
Re:More?! (Score:2)
"OMG! $2500 for 2GHz? WTF i can maek a 2GHz P4 for liek $250 d00d. ur messed up if u pay that much. Apple sucks LOL"
Sad but true.
Something still screwy with the test results (Score:1)
http://www.barefeats.com/g5sum.html
the dual 1.42 GHz G4 posted a time of 35 seconds while the dual 2.0 GHz G5 posted a time of 29 seconds. Barefeats EMPHATICALLY states the results are for Photoshop 7.01 without the G5 plug-in
On the recent comparison page, we see the exact same times (35 seconds vs. 29 seconds) for the MP-aware tests, leading one to assume that even this latest round of benchmarks were tested without the G5 plug-in, and thus don't give the re
Re:Something still screwy with the test results (Score:1)
Bus speed tests (Score:1)
- Jasen.