Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.

milo

macrumors 604
Sep 23, 2003
6,891
523
poppe said:
http://blogs.zdnet.com/Apple/?p=249

That was posted in another forum. Any truth behind this?

Considering it's a site that's rarely right, and he has no source? I wouldn't take it seriously. Especially since Apple would have known about something like this months ago and would have done the redesign already.
 

Apple Corps

macrumors 68030
Apr 26, 2003
2,575
542
California
milo said:
Considering it's a site that's rarely right, and he has no source? I wouldn't take it seriously. Especially since Apple would have known about something like this months ago and would have done the redesign already.

Good point - I'll keep my hopes up :)
 

gnasher729

Suspended
Nov 25, 2005
17,980
5,566
Apple Corps said:
Everything touts that Merom is 20% more powerful due to its design AND draws less power. However - the TDP suggests more cooling is needed in the laptop design. Are you reading this the same way I am?

Merom needs half the power (of a chip that needs twice as much power). Intel had published some comparisons where they compare the power consumption of Core 2 Duo chips with randomly chosen older Intel chips; I have never seen a power comparison with Yonah.

That said, a lot of power can be saved by turning the processor into lower power modes if you don't need all the Gigahertz available. In the benchmark test, that was likely not possible because the operating system most know exactly how to switch clock speed and voltage on the chip, and that is different from chip to chip. An operating system can't safely reduce the voltage of a Merom chip unless the programmers had real Merom chips to test it.

If you don't need 100% of the processors (which is the normal case), Merom will be able to do the same work at ten percent less clockspeed and save energy that way. Merom at 100% max speed takes more energy, but also does more work per second.
 

Apple Corps

macrumors 68030
Apr 26, 2003
2,575
542
California
gnasher729 - so if I go for highest performance I will have 20% "improvement" but draw more power (and create mor heat) than with the Yonah - thus the higher TDP - do I have it correct?
 

DavidC1

macrumors member
Jul 26, 2006
75
0
So.............. it sounds as if Merom REQUIRES a larger thermal dissipation design then Yonah. This makes it sound as if Apple will have to provide a better cooling system then is now in place - yes ??

Actually, there was a discussion in some other forums that showed that Apple's Core Duo systems, at least the laptops had poor thermal paste application. After some user re-pasted the thermal paste, the heat on the CPU went down dramatically. I don't know if
Apple fixed it now.

Everything touts that Merom is 20% more powerful due to its design AND draws less power. However - the TDP suggests more cooling is needed in the laptop design. Are you reading this the same way I am?

It should be very hard for Merom to reach its TDP. The problem with the Netburst based CPUs was that they had poor thermal/load throttling implementation. In fact, the same Enhanced Intel SpeedStep Technology on the Pentium M/Core Duo, is much superior to the Netburst's EIST. Real world tests showed that(http://www.xbitlabs.com/articles/cpu/display/pentiumm-780_9.html), Pentium M doesn't exceed TDP of 27W, while Pentium 4/Pentium D's with poor EIST/TM2 greatly exceed TDP.

Core 2 Duo has more finely grained power management, which Intel calls it Ultra Fine Grain power management. Core 2 Duo can turn of parts of the functional units(meaning FPU may not be in full power), while Core Duo only has control over the entire functional unit. It may turn out Core 2 Duo runs as cool as Core Duo, or even cooler.

Due to the fact that the lowest multiplier for Pentium M didn't change ever since the Banias Pentium M with 400MHz bus, the lowest idle speed increased everytime the CPUs got a new, faster FSB. It's not a problem now, because Merom is compatible with Napa platform.

It will become a problem with Santa Rosa, which brings a chipset with much higher TDP(70-80% higher), and higher idle clock speeds(1.2GHz). Hopefully the power management improves in Santa Rosa to offset the higher power consumption.
 

Apple Corps

macrumors 68030
Apr 26, 2003
2,575
542
California
DavidC1 - the thermal paste application issue has been "hotly" debated (my pun for the day).

The best concensus that I could derive after reading all of the posts is maybe a 2 degree C reduction. There was an initial report of huge and dramatic improvement by one person - but that appears to have given way to little improvement seen by most.

Thanks for your detailed response. I suspect that either the Yonah or Merom will meet my needs. Considering the issues with this revision, would you go for a Yonah based MBP in a few weeks or Merom it in early September if they are available then? Reliability is of more concern to me - either will have the necessary power.
 

manic

macrumors regular
May 29, 2006
103
0
I also feel that the benchmarks by PC Per was a bit poor. I realize that the room for improvement from yonah to merom is vastly inferior than that of P4 to Conroe, but I do feel that, perhaps, a few enhancements of the merom architecture might have been overlooked. Im not sure if those benchs were either optimized, could capture its 64bit features (whatever those may be) or even SSE4, which is supposed to dramatically enhance video conversion (h.264 in particular).

all im saying is, hold your horses, lets see some other benches to know for sure. Obviously i havent bought my mactel yet, but I think those who have have made a great deal nonetheless.
 

poppe

macrumors 68020
Apr 29, 2006
2,248
53
Woodland Hills
dailo said:
http://news.com.com/1606-2_3-6100051.html?part=rss&tag=6100051&subj=news

Intel claims Core Duo 2 can last an flight across the country doing normal tasks! If this is true, I might have to sell my MBP to get the new ones with Core Duo 2.

I saw that video and felt so good about waiting this whole summer with out a Mac...

Talk about brutal. I'v been wanting a Mac for a year now. And almost bought a Powerbook in December but said no I'll wait because of intel and not needing it. 17 " came out with my credit card, but I said no... Maybe wait till school starts perhaps something better will come along...

Oh man... Makes me excited thinking about my Nem merom coming to my house delivered in a sweet looking box... And having a 5 hour batter life laptop...
 

DavidC1

macrumors member
Jul 26, 2006
75
0
DavidC1 - the thermal paste application issue has been "hotly" debated (my pun for the day).

The best concensus that I could derive after reading all of the posts is maybe a 2 degree C reduction. There was an initial report of huge and dramatic improvement by one person - but that appears to have given way to little improvement seen by most.

Ok, I guess I should have looked at it beyond the first part. I don't know, as differences in cooling environment can make so much difference, the cooling system on the laptop should be worried more than the CPU itself.

Thanks for your detailed response. I suspect that either the Yonah or Merom will meet my needs. Considering the issues with this revision, would you go for a Yonah based MBP in a few weeks or Merom it in early September if they are available then? Reliability is of more concern to me - either will have the necessary power.

Reliability, hmm. Unless there's a problem with the chip, I doubt reliability will be different with Core 2 Duo(Merom). It may be tied to thermals again.
 

gnasher729

Suspended
Nov 25, 2005
17,980
5,566
risc said:
Good read thanks for the link. It looks like the Merom will be great for the scientists / mathmeticians and not make much difference to everyone else. I can't wait to see some real world results. It's weird to see the battery life go down with the Merom I can't say I was expecting that.

Battery life going down isn't unexpected at all if you read through the article. The put a Merom chip into a laptop, and the laptop didn't recognise the chip properly, basically treating it as a generic chip. Lots of things that are needed for saving battery power, like changing clockrate and voltage, require very detailed knowledge of the chip. If I were the programmer that was responsible for controlling the power savings mode, I wouldn't dare doing anything with a chip that is not recognised.
 

gnasher729

Suspended
Nov 25, 2005
17,980
5,566
DavidC1 said:
Core 2 Duo has more finely grained power management, which Intel calls it Ultra Fine Grain power management. Core 2 Duo can turn of parts of the functional units(meaning FPU may not be in full power), while Core Duo only has control over the entire functional unit. It may turn out Core 2 Duo runs as cool as Core Duo, or even cooler.

That would mean that I could run the H.264 encoder in Handbrake, having two CPUs running full throttle at 100 percent CPU usage, and still be saving power because the FPUs are turned off! That is excellent.
 

gnasher729

Suspended
Nov 25, 2005
17,980
5,566
Apple Corps said:
gnasher729 - so if I go for highest performance I will have 20% "improvement" but draw more power (and create mor heat) than with the Yonah - thus the higher TDP - do I have it correct?

If you have a really CPU intensive task, where you are waiting for the computer, that task should be finished 20 percent quicker. The CPU might use a bit more energy each second, but it would still use much less energy in total to perform the task. And of course the CPU is not the only thing in your computer, so you save more energy again. So as an extreme example, a Merom MacBook might be finished with a job after 2h 50m and five minutes later the battery runs out, while on a Yonah MacBook the battery lasts three hours, but isn't enough to finish the task.
 
Register on MacRumors! This sidebar will go away, and you'll see fewer ads.