Inside iPod Video, iMac G5, and PowerMac G5

[MacRumors]

Arstechnica provides an extensive review of the new 5G (Video) iPod. The review covers form factor, video playback, applications, and a dissection of the video iPod:

Taking this unit apart is a little bit different than previous models due to the changes in the housing design. Opening it was more akin to the nano and can be accomplished using a flat, hard tool like a screwdriver.

Meanwhile, Kodawarisan has posted images from an iMac G5 dissection. The iMac G5 was introduced during the October 12th media event alongside the Video iPod. The new iMac features a thinner case and built in iSight.

Finally, the first benchmarks comparing the new Dual Core PowerMacs (2.0 and 2.3) to the old Dual Processor 2.5GHz PowerMac. Note: benchmarks numbers from the Quad (Dual Dual-Core) PowerMac are not yet available. The benchmarks show that the Dual-Core processor (970MP) Macs have have comparable performance to the old Dual Processor (970) PowerMac even at lower gigahertz speeds. The Dual-Core 2.3GHz slightly edges out the Dual-Processor 2.5GHz PowerMac in the performance score totals.

 

[nagromme]

Always nice to dee details from Ars.

What's with apple and "flat hard tools"? It's like caveman-style hardware service

Lack of screws sure looks slick, though.

 

[longofest]

The benchmarks show that the Dual-Core processor (970MP) Macs have have comparable performance to the old Dual Processor (970) PowerMac even at lower gigahertz speeds. The Dual-Core 2.3GHz slightly edges out the Dual-Processor 2.5GHz PowerMac in the performance score totals.

Makes sense. As I have posted previously, dual-core can actually be better than dual-processor because of the benefits of cache-snooping, etc. Also, system architecture improvements such as DDR2 and PCI-e probably doesn't hurt

Edit: After a close look at the benchmarks, it appears as though the dual-core systems only beat out the 2.5Ghz dual-proc system because of memory-intensive tasks. The thread/CPU intensive tasks were won by the 2.5 Ghz model. So, cache-snooping might not add quite as much as I previously thought.

 

[Lollypop]

Am I reading Ars right... is the iPod videos battery replaceable? That’s the big thing I had against the nano.

I heard that DDR2 has a higher latency despite higher bandwidth.. how would this effect latency sensitive applications in comparison to older DDR Macs.

 

[~Shard~]

Great to see this kind of information, it's the kind of stuff I was waiting to find out! The dissection is cool (better someone else's Mac than my own! ) and the initial benchmarks are great to see as well. Can't wait for the quad benchmarks to come in, just to see where that beast really stands in relation to the other systems.

 

[Spock]

New iPods, iMacs, Powerbooks, POWERMACS I dont want this month to end! I cant wait to get my iPod!!

 

[longofest]

Anyone else notice that ALL of Apple's benchmarks were related to PREVIOUS PPC systems?

Smart move. A) it gives a good incentive for people like me to upgrade. B) it isn't going to come bite them in the arse when they move to Intel probably next revision.

 

[rosalindavenue]

What's up with the negative reviews of this story? These links are great; what could possibly make someone take the time to review them as "negative?" The Ars Technica review is as fine an example of tech journalism as I have ever seen.

 

[Lollypop]

What's up with the negative reviews of this story? These links are great; what could possibly make someone take the time to review them as "negative?" The Ars Technica review is as fine an example of tech journalism as I have ever seen.

Sometimes Ars does feel a little off but in general they do very good reviews!

Interesting though, wonder why Apple isnt/cant utilize the full capabilities of the Broadcom chip in the iPod?

 

[Hattig]

Anandtech did a review a month or two back that came to the conclusion that the PowerMac's memory controller had high latency which was letting the system down.

I suspect that the new memory controller has significantly less latency in the design, so whilst it is using higher latency DDR2 memory, the overall performance is increased. Hence in memory intensive tasks the new PowerMacs perform better.

In fact, I'm sure I saw on Apples site last night a reference to 'lower latency memory controller' ... wonder if I can find it again.

 

[Lacero]

w00t! They take it apart.

Definite read.

 

[Loge]

The review covers form factor, video playback, applications, and a dissection of the video iPod:

But not, it seems, sound quality.

 

[Chundles]

Sometimes Ars does feel a little off but in general they do very good reviews!

Interesting though, wonder why Apple isnt/cant utilize the full capabilities of the Broadcom chip in the iPod?

They will, but in a higher-priced full Video iPod designed around widescreen movies etc. They need to differentiate the products. Reading the review from ars just makes me want it more, CARN CHRISTMAS!!

 

[puckhead193]

Well this article answered my questions about it, the with of it... i'm glad idon't have to worry about with. Now i want to know a direct comparison between the 30 and 60, i'll prob get the 60 cause of the longer battery life...

 

[Lacero]

Interesting though, wonder why Apple isnt/cant utilize the full capabilities of the Broadcom chip in the iPod?

They probably can with an iPod software update to the firmware. Apple's holding it off for the next revision... so they can make more money on upgrades. Can't give consumers everything at once.

 

[Renegate]

Sometimes Ars does feel a little off but in general they do very good reviews!

Interesting though, wonder why Apple isnt/cant utilize the full capabilities of the Broadcom chip in the iPod?

What exactly are the capabilities?

 

[~Shard~]

They probably can with an iPod software update to the firmware. Apple's holding it off for the next revision... so they can make more money on upgrades. Can't give consumers everything at once.

Yep, that's usually the way it works. It isn't always about technology, sometimes marketing/business decisions have a nasty way of creeping in there...

 

[bentley]

What exactly are the capabilities?

video at 640x480, it's currently set at 320x240

 

[g4cubed]

Glad to see these benchmarks out so fast. But I'm waiting to see the quads marks before I decide which to buy.

So hurry up already.

 

[bowzer]

hmm.. is this new to the ipod 5th gen, the ability to record at 44.1khz in stereo?

i thought it was really low quality on the old ones.
i'd buy this in a second if it could record line in at 44.1khz stereo quality, i'd use it to record mix sets from my turntables.

 

[MrSugar]

hmm.. is this new to the ipod 5th gen, the ability to record at 44.1khz in stereo?

i thought it was really low quality on the old ones.
i'd buy this in a second if it could record line in at 44.1khz stereo quality, i'd use it to record mix sets from my turntables.

Well sir, then I suggest you buy.

From Apple's tech specs of the iPod

"Low (22.05 KHz, mono)
High (44.1 KHz, stereo)"

 

[bowzer]

Well the thing is, would it also be good music recording quality, because it could be great for recording voice, but suck for music.

Anyone know if you need buy some stupid 3rd party thing from griffin or belkin to get recording to work, or is it just as simple as plugging my audio source in, and pressing record/stop?
Because that would be nice.

 

[amac4me]

Makes sense. As I have posted previously, dual-core can actually be better than dual-processor because of the benefits of cache-snooping, etc. Also, system architecture improvements such as DDR2 and PCI-e probably doesn't hurt

Edit: After a close look at the benchmarks, it appears as though the dual-core systems only beat out the 2.5Ghz dual-proc system because of memory-intensive tasks. The thread/CPU intensive tasks were won by the 2.5 Ghz model. So, cache-snooping might not add quite as much as I previously thought.

As an owner of a Dual 2.5 system that I purchased in September 2004, I feel somewhat better that these news systems don't blow away my system. I wonder what the dual dual stats will be

 

[lilrabbit129]

What I really want to see are clock vs clock reviews. A Dual 2.3 vs a Dual-Core 2.3, Dual 2.0 vs. Dual-Core 2.0.

Also a comparison between the Quad 2.5 vs. Dual 2.5 and Dual 2.7 would be nice.

On the Ars review, the 2.5 beat out both systems in CPU and thread tests. The CPU I was expecting, but I thought the dual-cores would beat it out on the thread tests with the supposed benefits of dual-cores vs. dual-procs.

 

[Dont Hurt Me]

Anandtech did a review a month or two back that came to the conclusion that the PowerMac's memory controller had high latency which was letting the system down.

I suspect that the new memory controller has significantly less latency in the design, so whilst it is using higher latency DDR2 memory, the overall performance is increased. Hence in memory intensive tasks the new PowerMacs perform better.

In fact, I'm sure I saw on Apples site last night a reference to 'lower latency memory controller' ... wonder if I can find it again.

Thanks for this post. I think its at the heart of why these machines are faster then the older models. Something has never been right with that G5 controller. Well at least they solved it before making the jump.