MDD Motherboard "Clocking Down"?

[MatthewLTL]

So my logicboard from jbarley arrived today. I installed it and things seem fine however i ran into some hiccups fixed them but still have a question

All I did:

Installed 2GB RAM
Installed Battery
Installed Board
Installed 133MHz 1GHz CPU (From old board)
Installed Second Drive Cage and moved the main drives back to the ATA100 bus.

Booted the system to get the crash window.
Tried to book Live CD Lubuntu - nothing happened stopped at black screen
Tried to boot Leopard on the main HDD (Loaded and froze on the spinning Apple)
Tried to boot the Leopard USB Installer - Just sat at the with the spinner.

Held down ALT key for boot menu selected the main HDD and held the keys down to boot safe mode.

successfully booted into same mode and ran a disk verify on main HDD - it passed.

Rebooted now the system boots like normal


Now the question:

According to the Xbench results this was the motherboard from a DP 1.25GHz MDD. According to Intell, using my 133MHz CPU in a 167MHz board would clock down my CPU. However, it didn't "About This Mac" still shows the CPU as 1GHz. System Profiler shows the FSB being 133MHz.

This can only mean 1 of 2 things:

jbarley was wrong (which happens) and this was not a 167FSB board OR
Intell was wrong and the system Clocked down the FSB to accomidate the slower FSB CPU.

any thoughts?

 

[Intell]

Make sure you're using PC2700/PC3200 memory and not PC2100. Slower memory will hamper the bus speed.

 

[MatthewLTL]

Make sure you're using PC2700/PC3200 memory and not PC2100. Slower memory will hamper the bus speed.

all four stick register as PC2700 in the system profiler (even though all the ram is actually PC3200)

 

[jbarley]

jbarley was wrong (which happens)

Let me be quite clear about this, "it did not happen". I've only ever had, hell for that matter even seen one MDD Mac, and the xBench results I emailed to you were from that computer.
Here is an old link to where I was trying to give away the dual 1.25 CPU card from that same MDD.
I'm really pissed that you would suggest that I would mislead as to what that Logic board was, or even worse that I was too stupid to be able to figure out what it was.

BTW, I must have missed your "thank you" message.

 

[A.Goldberg]

Let me be quite clear about this, "it did not happen". I've only ever had, hell for that matter even seen one MDD Mac, and the xBench results I emailed to you were from that computer.
Here is an old link to where I was trying to give away the dual 1.25 CPU card from that same MDD.
I'm really pissed that you would suggest that I would mislead as to what that Logic board was, or even worse that I was too stupid to be able to figure out what it was.

BTW, I must have missed your "thank you" message.

I'm not sure if he's trying to criticize the fact that either you or Intell was wrong or figure out the perceived anomaly, or both. I suppose ultimately the OP should be happy that his computer is at running faster than anticipated, or in this case at all. If it wasn't for the replacement part I believe based on the threads all his computers- MDD, eMac, and HP would all be dead in the water. I assume you gave him the logic board for free? Sucks when go out of your way for people and they don't reciprocate or even thank you.

According to the Xbench results this was the motherboard from a DP 1.25GHz MDD. According to Intell, using my 133MHz CPU in a 167MHz board would clock down my CPU. However, it didn't "About This Mac" still shows the CPU as 1GHz. System Profiler shows the FSB being 133MHz.

Does the System Profiler just tell you the factory specs of the hardware that you have installed OR will it automatically update the new calculated speed of the the processor based off the logic board bus speed? I have zero experience with processor swaps this but it's just a thought.

So the computer is now booting normally without having to be in safe mode?

Edit: Matt- Did you personally run XBench on your MDD with the new board installed to see what results it gives out?

 

[MatthewLTL]

Let me be quite clear about this, "it did not happen". I've only ever had, hell for that matter even seen one MDD Mac, and the xBench results I emailed to you were from that computer.
Here is an old link to where I was trying to give away the dual 1.25 CPU card from that same MDD.
I'm really pissed that you would suggest that I would mislead as to what that Logic board was, or even worse that I was too stupid to be able to figure out what it was.

BTW, I must have missed your "thank you" message.

hey i was not meaning to offend you. It was not even about that. I was told the CPU would clock down on this board and it did not so THAT is what i was questioning. I am still working out the bugs i have not gotten to thanking you in a PM yet. "Board Swapping" isn't as simple as i thought and i have to fiddle around to get it to work.

 

[ctmpkmlec4]

hey i was not meaning to offend you. It was not even about that. I was told the CPU would clock down on this board and it did not so THAT is what i was questioning. I am still working out the bugs i have not gotten to thanking you in a PM yet. "Board Swapping" isn't as simple as i thought and i have to fiddle around to get it to work.

...

 

[MatthewLTL]

I'm not sure if he's trying to criticize the fact that either you or Intell was wrong or figure out the perceived anomaly, or both. I suppose ultimately the OP should be happy that his computer is at running faster than anticipated, or in this case at all. If it wasn't for the replacement part I believe based on the threads all his computers- MDD, eMac, and HP would all be dead in the water. I assume you gave him the logic board for free? Sucks when go out of your way for people and they don't reciprocate or even thank you.

Does the System Profiler just tell you the factory specs of the hardware that you have installed OR will it automatically update the new calculated speed of the the processor based off the logic board bus speed? I have zero experience with processor swaps this but it's just a thought.

To the OP: So the computer is now booting normally without having to be in safe mode?

Yes, It's booting normally. I have my ATA100 bus back so now i can run 4 HDDs again. The system is still a bit buggy though. Apparently OS X doesn't like when it has a board swap. The system itself is running stable apps on the other hand don't seem to like the swap too well.

As far as system profiler goes, I believe it is updated, It had no trouble at all showing my 2GB of RAM. I was not trying to criticize anyone. I am just trying to figure it out that is all. the Xbench jbarley provided shows it is a 167MHz board however the information Intell gave me says that putting my 133MHz CPU in a 167MHz board would clock down the CPU, It didn't. THAT is what this thread was about. It is not pointing fingers at anyone.

 

[A.Goldberg]

You could also try going into the terminal and typing in
sysctl hw.cpufrequency

I remember this command from long ago. The information it returns is kind of odd.

Explanation of sysctl hw.cpufrequency and what CPUfrequency does…
Posted on 11/19/2003 by C.K. Sample III
My friend, Jay Savage, who understands the BSD underbelly of OS X quite a bit better than I, explains what sysctl hw.cpufrequency does:

The poster at version tracker is correct that sysctl doesn’t return a real number. It returns a constant passed to the system at system start-up (and possibly wake-up, I’m not sure) by the PMU. The kernel and PMU will modify the actual speed on-the-fly based on load numbers and energy-saver setting in the pref pane, and sysctl will not return that number, you need to actually pool the processor to find out what it’s doing at any given time…this is what the bogomips calculated a Linux start-up are all about.

For your purposes, though, the real numbers don’t matter, because the bug in question involves the PMU essentially forgetting what the real speed is and setting the max Mhz artificially low. Effective clock speed may well often be below the number reported by sysctl (a kind of variable clock speed is what makes mobile chips possible: why waste power if your load numbers are below 1?), but it will never be above the number reported by sysctl. So if sysctl is reporting 400Mhz for a 800Mhz processor, you know something is seriously wrong. Even under full load, the PMU will not allow the CPU speed up past 400Mhz, which means you’re never getting more than 50% out of your hardware. Get that back up to 867, and you can theoretically get 867Mhz out of your chip. You won’t get that all the time, especially on battery power, but better and actual 400 out of 800 than an actual 200 out of 400.[bold emphasis mine]

So, the CPUfrequency scripts work fine for what they are meant to do, which is simply to report the maximum possible number (speed in hertz) your computer currently thinks in the processor is capable of achieving. If it reports back 5333333334 and you have an 867mhz chip, as happened with me earlier this week, then your PMU needs to be reset, so that the computer will realize that there is more processing power available there, should it be needed…

In this case, maybe see what resetting the PMU does? Just a thought.

I would also give Xbench a shot and see what numbers you come up with.

 

[MatthewLTL]

You could also try going into the terminal and typing in
sysctl hw.cpufrequency

I remember this command from long ago. The information it returns is kind of odd.



In this case, maybe see what resetting the PMU does? Just a thought.

I would also give Xbench a shot and see what numbers you come up with.

that command returned this: hw.cpufrequency: 999999997

 

[A.Goldberg]

Xbench jbarley provided shows it is a 167MHz board however the information Intell gave me says that putting my 133MHz CPU in a 167MHz board would clock down the CPU, It didn't. THAT is what this thread was about. It is not pointing fingers at anyone.

That's why you should run Xbench yourself and see what the results give you, now that you have the faster logical board and slower CPU paired.

By comparison of your 1ghz = 999999997

My 2.4Ghz rMBP = 2400000000
My 1.8Ghz G5 = 1800000000

I'm not sure if that -3 difference vs 1Ghz is indicative of anything. That article I posted states that a lower speed may indicate the need for a PMU reset. My line of logic would be the if Intell is correct, a -3 slower processing speed reflects the use of the faster logic board, though it seems like too small of a difference compared to what others were estimating. That's why I'd reset the PMU and the test again. Maybe someone else can comment if my line of reasoning/testing method makes sense here.

After the PMU reset, like I mentioned do the Xbench again to compare your results to JBarley. I suppose that would be the ultimate way of getting a more definitive answer of who is right and wrong here, if that's even the case.

 

[MatthewLTL]

That's why you should run Xbench yourself and see what the results give you, now that you have the faster logical board and slower CPU paired.

According to Xbench the CPU frequency is 134MHz still 1GHz with 1MB Level 3 Cache. It should be less than 1GHz according to Intell. So pretty much it's now about figuring out WHAT is slowing it down, Is the CPU clocking down the Motherboard? That seems the only logical thing. Because it is a FW800 board and the MDD it came out of had 1.25GHz DPs so that would mean 167MHz....

 

[A.Goldberg]

I am still working out the bugs i have not gotten to thanking you in a PM yet. "Board Swapping" isn't as simple as i thought and i have to fiddle around to get it to work.

If you have time to work on the project and write up this thread. You have time to send a thank you to the dude.

As far as system profiler goes, I believe it is updated, It had no trouble at all showing my 2GB of RAM.

The processor is different than the ram though

----------

According to Xbench the CPU frequency is 134MHz still 1GHz with 1MB Level 3 Cache. It should be less than 1GHz according to Intell. So pretty much it's now about figuring out WHAT is slowing it down, Is the CPU clocking down the Motherboard? That seems the only logical thing. Because it is a FW800 board and the MDD it came out of had 1.25GHz DPs so that would mean 167MHz....

My logical line of thinking would say that the logical board is limited by the speed of the CPU. If the 1Ghz CPU is at 133mhz, then the results you have would make sense even though the logic board is 167mhz. You're asking what's slowing it down- based on Intel's information, assumed to be true, the question would be it would be what is speeding it up? We would expect the processor to be slower with the faster logic board bus now in place.

I understand Intell is quite the guru and I admit am just working off my limited understanding of computers and mostly logic (though logic can often be counterintuitive). I trust his advice way more than mine. He does this for a living, I'm a pharmacist. Perhaps in this specific case or in this specific architecture the CPU is the limiting factor and not the logic board. Who knows? Sometimes things in life are just a mystery but you seem driven by curiosity. You can always do research on your own on the internet or consult other experts. On the plus side it sounds like you have your main machine up an running again which was the real goal here.

 

[MatthewLTL]

If you have time to work on the project and write up this thread. You have time to send a thank you to the dude.



The processor is different than the ram though

i can try a CUDA reset.. since the motherboard was not in a computer and with no battery you'd think that would make everything reset but maybe not

 

[MatthewLTL]

If you have time to work on the project and write up this thread. You have time to send a thank you to the dude.



The processor is different than the ram though

----------



My logical line of thinking would say that the logical board is limited by the speed of the CPU. If the 1Ghz CPU is at 133mhz, then the results you have would make sense even though the logic board is 167mhz. You're asking what's slowing it down- based on Intel's information, assumed to be true, the question would be it would be what is speeding it up? We would expect the processor to be slower with the faster logic board bus now in place.

I understand Intell is quite the guru and I admit am just working off my limited understanding of computers and mostly logic (though logic can often be counterintuitive). I trust his advice way more than mine. He does this for a living, I'm a pharmacist. Perhaps in this specific case or in this specific architecture the CPU is the limiting factor and not the logic board. Who knows? Sometimes things in life are just a mystery but you seem driven by curiosity. You can always do research on your own on the internet or consult other experts. On the plus side it sounds like you have your main machine up an running again which was the real goal here.

Yes. Phase 1 of the upgrade stage was a GPU but if i can get a motherboard for the sole cost of shipping I ain't gonna pass it up. I actually surprised it Chimed. Lots of mixed answers about this upgrade for people on here. 1st i was told the CPU would overclock to 1.25GHz than i was told it wouldn't boot at all than Intell said it would clock down, Surprisingly, STAYING the same speed was the surprise I was expecting to see something in the 867MHz-970MHz or so. Now is it possible to overclock the CPU itself make it a 167MHz CPU?

 

[A.Goldberg]

Yes. Phase 1 of the upgrade stage was a GPU but if i can get a motherboard for the sole cost of shipping I ain't gonna pass it up. I actually surprised it Chimed. Lots of mixed answers about this upgrade for people on here. 1st i was told the CPU would overclock to 1.25GHz than i was told it wouldn't boot at all than Intell said it would clock down, Surprisingly, STAYING the same speed was the surprise I was expecting to see something in the 867MHz-970MHz or so. Now is it possible to overclock the CPU itself make it a 167MHz CPU?

I meant research independent outside of MacRumors PPC followers, like talking to computer scientists/engineers and not just tinkerers and enthusiasts. People who better understand the engineering behind the technology. Or perhaps reading websites or materials, building your own knowledge on your own, rather than just posting questions.

I'm not sure, but it would seem much easier and much much safer just to find a 1.25GHz 166MHz CPU to throw in there. I can't imagine it would cost much either.

 

[MatthewLTL]

I'm not sure, but it would seem much easier and much much safer just to find a 1.25GHz 166MHz CPU to throw in there. I can't imagine it would cost much either.

Any idea if there will be a notiicable increase in preformance with a 1.25GHz CPU?

 

[A.Goldberg]

Any idea if there will be a notiicable increase in preformance with a 1.25GHz CPU?

I think the speed increase would be quite negligible and not worth investing in unless you get the processor for nothing or next to nothing. You would notice a real difference if you switched to dual processors.

But you seem to do things just for the sake of doing it.

 

[ctmpkmlec4]

Here you go. Again, this is not my listing.

 

[MatthewLTL]

I think the speed increase would be quite negligible and not worth investing in unless you get the processor for nothing or next to nothing. You would notice a real difference if you switched to dual processors.

But you seem to do things just for the sake of doing it.

would you see the preformace jumping from say a DP 1.25 to a DP 1.33?

 

[bunnspecial]

Many folks consider to be the single 1.25ghz to be the fastest stock OS 9 computer available. OS 9 itself does not make use of multiple processors, although specific applications running within OS 9 can(Photoshop 7 is one that comes immediately to mind) Of course, that would be a moot point since the FW800 isn't readily bootable into OS 9(there are workarounds).

A single 1.25 should get your system bus up to 167 mhz, and will also give you 2mb of L3 cache. Both of these will have a noticeable impact on performance. For OS X use, though, I'd choose a dual 1ghz.

FWIW, below, from left to right, are Geekbench scores from a dual 1ghz Quicksilver, dual 1ghz MDD, and single 1.25ghz MDD. This particular Quicksilver is somewhat crippled as the L3 Cache never worked-I think its replacement(with working L3 cache) gets in the upper 800s or lower 900s-or in other words pretty similar to the dual 1ghz MDD.

 

[A.Goldberg]

Many folks consider to be the single 1.25ghz to be the fastest stock OS 9 computer available. OS 9 itself does not make use of multiple processors, although specific applications running within OS 9 can(Photoshop 7 is one that comes immediately to mind) Of course, that would be a moot point since the FW800 isn't readily bootable into OS 9(there are workarounds).

A single 1.25 should get your system bus up to 167 mhz, and will also give you 2mb of L3 cache. Both of these will have a noticeable impact on performance. For OS X use, though, I'd choose a dual 1ghz.

FWIW, below, from left to right, are Geekbench scores from a dual 1ghz Quicksilver, dual 1ghz MDD, and single 1.25ghz MDD. This particular Quicksilver is somewhat crippled as the L3 Cache never worked-I think its replacement(with working L3 cache) gets in the upper 800s or lower 900s-or in other words pretty similar to the dual 1ghz MDD.

Take it from Bunn, he has every PPC model ever made!!

So $19 w/shipping for the upgrade. Plus the cost of thermal paste if you have it. On the one hand, a G5 can be acquired for what, $70? ...which isn't much compared to if you keep making small, incremental improvements on your other computer(s). On the other hand, you may prefer the MDD over the G5 and $19 isn't a whole lot of money.

 

[redheeler]

That's why you should run Xbench yourself and see what the results give you, now that you have the faster logical board and slower CPU paired.

By comparison of your 1ghz = 999999997

My 2.4Ghz rMBP = 2400000000
My 1.8Ghz G5 = 1800000000

I'm not sure if that -3 difference vs 1Ghz is indicative of anything. That article I posted states that a lower speed may indicate the need for a PMU reset. My line of logic would be the if Intell is correct, a -3 slower processing speed reflects the use of the faster logic board, though it seems like too small of a difference compared to what others were estimating. That's why I'd reset the PMU and the test again. Maybe someone else can comment if my line of reasoning/testing method makes sense here.

After the PMU reset, like I mentioned do the Xbench again to compare your results to JBarley. I suppose that would be the ultimate way of getting a more definitive answer of who is right and wrong here, if that's even the case.

If it still matters, from my 1.0 GHz eMac: hw.cpufrequency: 999999997

 

[desantii]

No real big difference from what you what which is not spending much your best bang for your buck is a DUal 1.25 which is a relatively cheap upgrade Diminishing returns after that

would you see the preformace jumping from say a DP 1.25 to a DP 1.33?

 

[MatthewLTL]

Many folks consider to be the single 1.25ghz to be the fastest stock OS 9 computer available. OS 9 itself does not make use of multiple processors, although specific applications running within OS 9 can(Photoshop 7 is one that comes immediately to mind) Of course, that would be a moot point since the FW800 isn't readily bootable into OS 9(there are workarounds).

A single 1.25 should get your system bus up to 167 mhz, and will also give you 2mb of L3 cache. Both of these will have a noticeable impact on performance. For OS X use, though, I'd choose a dual 1ghz.

FWIW, below, from left to right, are Geekbench scores from a dual 1ghz Quicksilver, dual 1ghz MDD, and single 1.25ghz MDD. This particular Quicksilver is somewhat crippled as the L3 Cache never worked-I think its replacement(with working L3 cache) gets in the upper 800s or lower 900s-or in other words pretty similar to the dual 1ghz MDD.

what about the DP 1.25, 1.33 and 1.42 compared to the DP 1.0?