Developer Transition Kit Benchmarks

[MacRumors]

Despite the restriction on publishing benchmarks, the the first Xbench benchmarks are trickling in.

Here is one XBench screenshot showing the developer 3.6GHz Pentium 4 PowerMac running Xbench under the Rosetta emulation. Screenshot

On the left is the Xbench under emulation and on the right is Xbench running on a Dual 2.7GHz PowerMac G5. This simply shows that Rosetta certainly works, but is hard to give us real performance comparisons until the final shipping machines come.

ThinkSecret notes that the Intel Mac scored well in both the Quartz graphics and OpenGL graphics tests "almost matching or exceeding dual-2.5GHz G5 score".

Readers should note, that while all PowerPC native instructions are translated/emulated, since the underlying Mac OS X is running natively on the Intel processor, system tools calls/APIs should enjoy native (not emulated) speeds.

 

[RichP]

Wow..that was quick..now we need some pictures of the machines!

 

[polyesterlester]

It would be awesome if someone would explain these two charts in plain english, so some of the less educated among us (including me) can understand everything in its entirety.

EDIT: I accidentally didn't read the Think Secret post. It explained things pretty well.

 

[barneygumble]

NDA's are for losers anyway.

 

[KC9AIC]

These should be improved by the fact that there are faster Intel processors in the works, coming at a quicker rate than faster PPC processors.

 

[javiercr]

is macrumors going to get in trouble for publishing this?

 

[mj_1903]

Wow..that was quick..now we need some pictures of the machines!

The machines are simply G5 cases.

 

[DHagan4755]

It's too early to tell.

 

[RichP]

The machines are simply G5 cases.

I should have specified, I meant the INSIDE of the machines; MB configuration, chipset, etc.

 

[Kelmon]

Well, it's definitely going to have to be a case of "wait and see". The initial benchmarks aren't promising but then we're not expecting to be using development machines in the same way that XBox360 owners won't be expecting a PowerMac when their console arrives. I'm definitely waiting to see what a shipping system will be equipped with before panicking.

One thing is for sure, the disparity in performance between a G5 and P4 in these benchmarks explains why Apple didn't want benchmarks to be posted.

 

[BWhaler]

This is annoying as all hell.

Now the rumors forever are going to be Macs on Intel are slow...blah, blah, bla.

And of course, none of this stuff--hardware, OS, software--has been optimized yet.

This is just noise in the system we don't need.

 

[Nermal]

NDA's are for losers anyway.

You have to sign an NDA in order to rent a kit, but I wonder whether you need to sign anything to use one onsite at WWDC. It's possible that no NDA is being broken here.

 

[richard5mith]

These are pretty invalid benchmarks because surely Xbench is being emulated. So if your benchmarking software is running under emulation, things aren't going to go well.

 

[whooleytoo]

The floating point and vecLib are the scary benchmarks in those pics - 25+ and 45+ times slower respectively. Very early days yet, so it's impossible to infer anything useful from these benchmarks.

 

[SPUY767]

Looks to me that the 2.7 pretty much wiped its ass with the intel machine. Graphics didn't suffer since they are mainly system level GPU calls. But everything else seemed to get hammered by the G5. This benchmark means bugger all however, because it's being emulated. Suffice to say, that the benchmarks fared better than they would if it was being run on PearPC.

 

[AmigoMac]

This is annoying as all hell.

Now the rumors forever are going to be Macs on Intel are slow...blah, blah, bla.

And of course, none of this stuff--hardware, OS, software--has been optimized yet.

This is just noise in the system we don't need.

Like the " Buy or wait?" threads!!!

Will VPC on OS X for Intel macs emulate Solitaire at full speed?

Let's wait for the "I won't buy this and will stay with my iBook" !!

 

[flater1975]

Not X86 benchmarks?

Nevermind. Reading is not my strongest point

 

[MacsRgr8]

On the left the x86, and on the right the Dual G5

Too bad we can't find out what hardware is used on the x86 side.
CPU = P4 3.6 GHz?
FSB = 800 MHz?
Grfx = Radeon X850?

Can't wait to get the first real pics of that machine

 

[abrooks]

I agree with most people here these Benchmarks are almost a waste of space, for one we don't have a complete list of Hardware on the x86 side and even the hardware that is listed is not up to a similar spec of the PPC side.

All in all its too early to tell, the software is not optimised and neither is the hardware I assume.

 

[efoto]

This is annoying as all hell.

Now the rumors forever are going to be Macs on Intel are slow...blah, blah, bla.

And of course, none of this stuff--hardware, OS, software--has been optimized yet.

This is just noise in the system we don't need.

Like the " Buy or wait?" threads!!!

Will VPC on OS X for Intel macs emulate Solitaire at full speed?

Let's wait for the "I won't buy this and will stay with my iBook" !!

Ha, too true. We are going to hear for ages that OSX on x86 chips is slower and that Rosetta doesn't work. "You can't believe that, even your benchmarks are emulated." I can hear the endless, worthless banter for PC users everywhere just starting to murmer louder and louder.

They will speak without knowing anything about it, and claim that since Apple switched to x86, we are using the same system but paying more for it. I think if anything it will be harder to switch someone now that it was two weeks ago (overlooking pre-WWDC purchases....since we all know how those go ). I'm confident now that SJ will bring us all through, I just hope that there is clear and concise evidence that makes things easier to describe to potential switchers.

 

[Dont Hurt Me]

2 G5s = 1 Intel or AMD. Nothing new here.

 

[efoto]

is macrumors going to get in trouble for publishing this?

I suppose there is a possibility, however since 'we' as the community of MacRumors didn't receive a Development Box and 'we' did not run the bench and 'we' did not publish the results, 'we' did not violate the NDA that 'we' never signed.

So in short, no.

 

[bushgreen]

maybe the G5 is just faster than a P4.

 

[wbiggart]

can anyone post a screenshot of the "About This Mac" dialog so that we can get a peek at what type of processor these intel boxes are running with.

i know that these are just developer test-beds, and that the processor is going to change between now and when these boxes hit the market, but it'd be interesting none the less.

 

[bellis1]

comparison

could we get a comparison of a mini or imac? i hope these developers are going to get a pay raise.