Can't you just overclock your 2.0 macbook?

[Germwise]

Anyone with more expertise, please correct me if I'm wrong and add your thoughts to this thread.

The way I understood computer chips was that chips from the same architecture of different speed ratings (2 ghz, 3ghz etc) were in fact identical chips.

From what I understood from overclocking many PCs was that during the manufacturing process there was a variation in the quality of the chip. Thus a chip of architecture X, that came out "flawless" would be rated at 3ghz, while one that maybe didn't do so well in quality control would get rated at 2ghz.

Sometimes though, the chip is locked at a lower speed when it is perfectly capable of reaching much higher speeds.

So in essence, aren't the 2.0ghz UMB and the 2.4 UMB the same chip wise? I know the FSB is the same so there must just be a difference in the multiplier

(chip speed is FSB*Multiplier so 1066*2.3=2.4)


Do we have access to changing the multiplier on macbooks to see if they will run stable at the higher speed? I imagine with an aluminum unibody, the slight increase in heat shouldn't make much of a difference.

I haven't done any overclocking in many years so please correct me but it seems like with a little tinkering you could run 2.0 ghz macbooks at 2.4 levels. Not that it makes that much of a diff.

 

[Toofan]

Typically, you should be able to overclock anything if you can reach it at the hardware level. I would assume (i'm not certain) with BSD acting as a lower level unix os in OSX, you should be able to do everything from overclocking to manually clearing your RAM. That being said, if your chip was meant to be over clocked, they would have sold it as a higher level chip. Often what Intel does is physically cut wiring in their chips to make them run at lower speeds, it's all about supply and demand. Just like an Intel triple core machine is actually a quad core with a bad core which has been disconnected.

Once again, I know you were just wondering, but I highly advise against overclocking a 2.0ghz chip in a laptop especially. The cooling just isn't there, and as you acknowledged, the benefits are laughably small. To be frank, the whole 2.0ghz vs 2.4ghz debate is all hokey. That 400mhz doesn't do much for anyone but video encoders in the long run.

 

[modular]

dude, i can just build my own laptop and make it twice as fast as these macbooks. it's so easy, i can't believe people just don't build their own laptops.

 

[flopticalcube]

Most Intel chips have their clock multipliers locked in hardware. The only way you could overclock them is to get into the (if you could) EFI and rev up the FSB, which has serious implications elsewhere in the laptop. Its just not doable.

 

[Germwise]

build your own laptop?

I've put together many a desktop.

But unless you are in the business of designing logic boards I doubt you can "build" your own laptop.


Of course you can buy a dell and make a hackingtosh but after buying leopard separately and being in fear that the next update will break your laptop isn't for me.

I also love the aluminum form factor. Call me a sucker but I think its important to like the machine you spend 5-16 hours a day on.

 

[pellets007]

Yeah, you could probably OC the MacBook's CPU, IGPU, and even the RAM. But you don't do that it laptops for fear of heat problems. The software is not on par with chips faster than 2.4Ghz nor are the fans I would assume. That said, I am not aware of any such program under OS X. But if you could try it out, you would be on the frontier I believe.

 

[NewMacbookPlz]

There are other differences between the P7350 and P8600 chips, more than just clock speed.

One big one is the lack of on-CPU virtualization with the P7350. For us OSX guys it doesn't really matter, but for Windows-7 it will mean the WindowsXP virtualization won't be able to run.

 

[Gregintosh]

The way I understood computer chips was that chips from the same architecture of different speed ratings (2 ghz, 3ghz etc) were in fact identical chips.

From what I understood from overclocking many PCs was that during the manufacturing process there was a variation in the quality of the chip. Thus a chip of architecture X, that came out "flawless" would be rated at 3ghz, while one that maybe didn't do so well in quality control would get rated at 2ghz.

That is absolute baloney. Most likely an "urban myth."

Given that Intel produces millions if not tens or hundreds of millions of processor a year, they certainly would not have the time or capacity to test each chip individually. Therefore, it would be impossible to "rate" their clock speeds based on individual testing or quality control, which could take a long time to produce conclusive results.

Second of all, Intel would not release chips that did not meet quality control standards period.

There are real differences between the chips, sometimes minor, but there are differences.

One time I heard an urban myth at a large electronics retail chain in the US that Intel was better because defective Intel chips get sold to AMD, which proceeds to sell them at lower speeds and prices.

I would chalk up this "if the processor doesn't pass Quality Control or is shown to be inferior, it will be just clocked down and sold anyway" to the same level of myth as "AMD's are just defective Intels"

I think its important to like the machine you spend 5-16 hours a day on.

That I agree with. I never understood people who get something ugly just because it is cheaper, especially when it comes to something that they will be staring at most of their day for the next year or two to come. I can understand desktops, because you can put it under your desk and get a nice looking monitor, but with laptops there just isn't an excuse.

And overclocking isn't risk-free. It also reduces the life of the chip. I am pretty sure it would also void any warranty you have, including AppleCare. Better safe than sorry.

 

[Germwise]

This isn't just something I've "heard" its through much research


this is from the wiki on overclocking

Factors allowing overclocking


CPU-Z screenshot demonstrating an E2140 overclocked to twice its rated frequency.
Overclockability arises in part due to the economics of the manufacturing processes of CPUs and other components. In most cases components with different rated clock speeds are manufactured by the same process, and tested after manufacture to determine their actual ratings. The clock speed that the component is rated for is at or below the speed at which the CPU has passed the manufacturer's functionality tests when operating in worst-case conditions (for example, the highest allowed temperature and lowest allowed supply voltage). Manufacturers must also leave additional margin for reasons discussed below. Sometimes manufacturers produce more high-performing parts than they can sell, so some are marked as medium-speed chips to be sold for medium prices. The performance of a given CPU stepping usually does not vary as widely as the marketing clock levels[citation needed].

 

[pellets007]

This isn't just something I've "heard" its through much research


this is from the wiki on overclocking

Read the last few words on that quote.

 

[Gregintosh]

Well if "citation needed" is your source, then I am sold. I just wish my professors would let me use that same source to fill in gaps in my research reports.

The article implies that Intel has no control over their supply. If Intel wants 500,000 2.66Ghz chips, the article implies they wouldn't be able to set their equipment to just create 500,000 of such chips, instead, they would have to manufacture possibly millions of chips and hope that enough of them rate at 2.66Ghz or higher. It just doesn't sound right.

I'm no engineer, but from what I know about business, manufacturing is usually a highly calculated, controlled, and predictable process.

I don't doubt that differences between chips might be small, and slower chips might be capable of higher speeds.

It is more plausible, however, that they keep a buffer zone before the maximum clock rating of the chip because a higher clock rating increases the chance of the chip becoming defective, beyond what the quality control team says is acceptable to keep consumers happy. This rating exercise would be done for every individual model, not for every individual chip. In reality, though, with chip design being a precision science down to almost the molecule level, I doubt the testing is used for anything other than simply verifying the predictions and decisions of the chip designers.

This buffer zone, then, is what is exploited by overclockers who overclock the chips at their own risk. Many get lucky and many control the effects of overclocking by installing non-stock, superior cooling systems to those the chips were designed to work with.

It would also still mean that the chips are in fact different so no, the 2.0 and 2.4 chips in macbooks are not the same chip (nor are they the same as higher clocked chips) with just a different setting for marketing purposes.

There is no conspiracy to keep processors slow, so if Intel was able to clock all their chips across the board higher while maintaining the same quality standard and low-defect rate, they would do so and leave AMD even further behind in the dust. You can overclock, and yes, Intel chips are capable of overclocking quite well in many cases, but you do so at your own risk and believe me, there is a risk.

 

[mosx]

There are other differences between the P7350 and P8600 chips, more than just clock speed.

One big one is the lack of on-CPU virtualization with the P7350. For us OSX guys it doesn't really matter, but for Windows-7 it will mean the WindowsXP virtualization won't be able to run.

Thats not true.

The P7350 in the MacBook DOES support virtualization.

SecureAble, Intel's own processor identifier software, SiSoft Sandra all state that it does support Intel's virtualization technology.

Theres also this guy http://ivoidwarranties.blogspot.com/2009/04/core-2-duo-p7350-vs-intel-vt-x.html

As well as others (not just me) who ran those tools I listed above under Windows at other forums and all confirmed that the P7350 in the MacBook does indeed support virtualization.

Anyone with a 2GHz unibody MacBook or the new white MacBook would either run that terminal command in that link above or run one of those tools I listed in Windows to confirm that the P7350 does support Intel's virtualization extensions.

 

[jav6454]

Thats not true.

The P7350 in the MacBook DOES support virtualization.

SecureAble, Intel's own processor identifier software, SiSoft Sandra all state that it does support Intel's virtualization technology.

Theres also this guy http://ivoidwarranties.blogspot.com/2009/04/core-2-duo-p7350-vs-intel-vt-x.html

As well as others (not just me) who ran those tools I listed above under Windows at other forums and all confirmed that the P7350 in the MacBook does indeed support virtualization.

Anyone with a 2GHz unibody MacBook or the new white MacBook would either run that terminal command in that link above or run one of those tools I listed in Windows to confirm that the P7350 does support Intel's virtualization extensions.

Wrong! P7350 does not support Intel VT, which is essential for Windows 7 to run in XP mode.

It is all over the internet and even Intel as acknowledged to it. Many laptop owner will be screwed as the P7350 is a very popular CPU among cheap laptop makers.

 

[mosx]

Wrong! P7350 does not support Intel VT, which is essential for Windows 7 to run in XP mode.

It is all over the internet and even Intel as acknowledged to it. Many laptop owner will be screwed as the P7350 is a very popular CPU among cheap laptop makers.

rofl I'll post a screen shot later tonight.

I love it when people tell you "NO! YOU'RE WRONG! YOU DID NOT SEE WHAT YOU SAW!"

 

[Eidorian]

Wrong! P7350 does not support Intel VT, which is essential for Windows 7 to run in XP mode.

It is all over the internet and even Intel as acknowledged to it. Many laptop owner will be screwed as the P7350 is a very popular CPU among cheap laptop makers.

I wonder why the P7350 keeps reporting VT support then. In before screenshots.

 

[jav6454]

rofl I'll post a screen shot later tonight.

I love it when people tell you "NO! YOU'RE WRONG! YOU DID NOT SEE WHAT YOU SAW!"

There Click Me, if you can't trust magazine and other sites, at least trust Intel, they know, they made the CPU.

BTW, screenshots mean nothing as they can be shoop'd anyways.

 

[Eidorian]

There Click Me, if you can't trust magazine and other sites, at least trust Intel, they know, they made the CPU.

BTW, screenshots mean nothing as they can be shoop'd anyways.

The screenshots won't be shopped but it is part of a much bigger mess.

 

[jav6454]

The screenshots won't be shopped but it is part of a much bigger mess.

Touche, but still, P7350 doesn't have Intel VT, and any screenshot saying so, won't make the CPU magically have VT.

 

[Eidorian]

Touche, but still, P7350 doesn't have Intel VT, and any screenshot saying so, won't make the CPU magically have VT.

It's a reporting problem based on how the OEM handles it and the BIOS.

 

[mosx]

There Click Me, if you can't trust magazine and other sites, at least trust Intel, they know, they made the CPU.

BTW, screenshots mean nothing as they can be shoop'd anyways.

Photoshopping screenshots. Right

All I did was underline the parts to show that you're wrong.

 

[mosx]

Touche, but still, P7350 doesn't have Intel VT, and any screenshot saying so, won't make the CPU magically have VT.

Don't forget the link I posted as well.

 

[jav6454]

Stick with a no, Intel says no, don't try to go against that, its the manufacturer for Christ sakes.

If it runs, great, but if it doesn't then don't say later you were warned.

 

[zunairryk]

Well if "citation needed" is your source, then I am sold. I just wish my professors would let me use that same source to fill in gaps in my research reports.

lol thanks for making me laugh

 

[mosx]

Stick with a no, Intel says no, don't try to go against that, its the manufacturer for Christ sakes.

If it runs, great, but if it doesn't then don't say later you were warned.

Okay you've got a number of things going against you.

First, you have no denial from Intel that the processor supports it.

Second, you have Intel's OWN software confirming that it DOES support it.

Third, you have two pieces of very highly respected software also confirming that support.

Fourth, you have a blog post showing OS X confirming the support as well as that user running software in OS X that REQUIRES VT/VMX support. It would NOT be possible for that user to run 64-bit Windows in a virtual machine without VT/VMX support.

Fifth, you can go to Ars forums and others where other users with 2GHz UniBody and white nvidia MacBooks also run the same software I did confirming the chips support.

So are you still going to say "no"?

Edit: I'm not saying that all P7350s support VT/VMX. It's obvious they don't. This is one of those features where its up to the OEM to support it. In fact, theres a thread over at Ars where an actual Intel employee posts talking about how its up to OEMs to enable support for features on the hardware they build. There are other posters in that thread also discussing how they've had BIOS updates from their system manufacturers disable features like VT/VMX on certain processors that previously had support.

But in this case, the P7350 in the unibody and plastic MacBook DOES support hardware virtualization.

Keep in mind that the Core Duos were not supposed to support the technology either. Obviously they don't support 64-bit technology. Intel and Apple both claimed it was disabled on the Core Duo Macs. But then Parallels came out and showed that it WAS supported.

 

[em500]

Stick with a no, Intel says no, don't try to go against that, its the manufacturer for Christ sakes.

If it runs, great, but if it doesn't then don't say later you were warned.

Have you ever considered the possibility that some of the bigger OEMs such as Apple can get slightly customized CPUs from Intel?