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

propower

macrumors 6502a
Jul 23, 2010
731
126
As an EE you do know that electrolytic capacitors have a rated life of a certain number of hours at a certain maximum temperature with a certain maximum ripple and that the life is very much affected by the operating temperature and ripple.

I know these factors very well :)! Just a few years ago AL Electrolytics were all 105degC. Now you can spec in 125 degC versions. I note on the ifixit board that all the caps next to the CPU look like tantalums and I am hoping are all 125degC guys. The Electrolytics on the top I would need to see the serial number. For longest life though you have my vote that something like 80% of rating is about as hard as you really want to run things. Anything over 100degC is a worry point and 110degC is a cliff no one wants to be near unless you totally know what you are doing.

I get why Apple chose this route, and it will probably work fine for the use and lifetime I expect. But I just don't like it :). Will certainly be looking at the new pro whenever it finally appears.

As a last point, I always thought there would be a market for repackaging the guts of the mini in a 2X sized box with great heatsinks, low noise fans and 60degC max proc temp with < 34dBA performance at full load. It would of course void everything but could probably be done for ~$200!
 

motrek

macrumors 68030
Sep 14, 2012
2,613
305
As an EE you do know that electrolytic capacitors have a rated life of a certain number of hours at a certain maximum temperature with a certain maximum ripple and that the life is very much affected by the operating temperature and ripple. The IBM T-series Thinkpads were "over engineered" and I know of some of these laptops still running 10 years later on original batteries..... (I ran a couple of them for 3 years almost daily at 100% and they are still running today 7 years later with the guy I sold them to).

Not sure why you think 7 years is a big achievement. I bought a cheap Compaq laptop in 2004 (i.e., 8 years old) which I eventually gave to my parents and they used it heavily until the LCD backlight broke just recently... as a computer, it still works fine, if you plug it into an external monitor. My grandmom has an original iMac from 1998, i.e., 14 years old, and it still works too, although the monitor had to be repaired a few years ago.

In my experience, solid state computer electronics will pretty much work indefinitely as long as it isn't affected by faulty capacitors from the early 2000s.

My prediction is that in 7 years nobody will remember all this discussion of Minis overheating and planned obsolescence because all the 2012 Minis will still be working fine.
 

motrek

macrumors 68030
Sep 14, 2012
2,613
305
I know these factors very well :)! Just a few years ago AL Electrolytics were all 105degC. Now you can spec in 125 degC versions. I note on the ifixit board that all the caps next to the CPU look like tantalums and I am hoping are all 125degC guys. The Electrolytics on the top I would need to see the serial number. For longest life though you have my vote that something like 80% of rating is about as hard as you really want to run things. Anything over 100degC is a worry point and 110degC is a cliff no one wants to be near unless you totally know what you are doing.

I get why Apple chose this route, and it will probably work fine for the use and lifetime I expect. But I just don't like it :). Will certainly be looking at the new pro whenever it finally appears.

As a last point, I always thought there would be a market for repackaging the guts of the mini in a 2X sized box with great heatsinks, low noise fans and 60degC max proc temp with < 34dBA performance at full load. It would of course void everything but could probably be done for ~$200!

Right now I'm using a hackintosh with a Scythe Mugen 2 heatsink (as big as a baby's head) with two 120mm Nexus fans running at 750 RPM. Other than those fans and a hard drive there are no moving parts, and my temperatures never get above ~55C, so you can see that I would be exactly the target demo for a Mini like that.

But that being said, being an engineer (my degree is in EE too), I'm sure you're familiar with how these products are tested. Apple probably did every possible kind of test, including accelerated aging, on a thousand Minis before ramping up production. People here think running Prime95 for a few hours is a stress test--I can guarantee that Apple ran Prime95 or similar software for *weeks* with every possible ambient temperature, etc.
 

dasx

macrumors 65816
Jun 18, 2012
1,107
18
Barcelona

philipma1957

macrumors 603
Original poster
Apr 13, 2010
6,367
251
Howell, New Jersey
Right now I'm using a hackintosh with a Scythe Mugen 2 heatsink (as big as a baby's head) with two 120mm Nexus fans running at 750 RPM. Other than those fans and a hard drive there are no moving parts, and my temperatures never get above ~55C, so you can see that I would be exactly the target demo for a Mini like that.

But that being said, being an engineer (my degree is in EE too), I'm sure you're familiar with how these products are tested. Apple probably did every possible kind of test, including accelerated aging, on a thousand Minis before ramping up production. People here think running Prime95 for a few hours is a stress test--I can guarantee that Apple ran Prime95 or similar software for *weeks* with every possible ambient temperature, etc.

based on 2011 server returns ( high rate of bad mobos) you are not correct.
 

motrek

macrumors 68030
Sep 14, 2012
2,613
305
based on 2011 server returns ( high rate of bad mobos) you are not correct.

C'mon Philip, you are always making these sorts of claims (Minis are designed to fail in 3 years, 2011 servers all got returned, etc.) which you would have no way of knowing unless you worked at Apple.

Can you substantiate anything you've been saying lately?
 

Snowcake

macrumors regular
May 18, 2010
187
0
C'mon Philip, you are always making these sorts of claims (Minis are designed to fail in 3 years, 2011 servers all got returned, etc.) which you would have no way of knowing unless you worked at Apple.

Can you substantiate anything you've been saying lately?

Yes, i do. But i am afraid that's all i can tell. I signed a NDA.
 

philipma1957

macrumors 603
Original poster
Apr 13, 2010
6,367
251
Howell, New Jersey
all this reminds me of the 2009 27 inch iMacs .. I complained I bi-atched about the encased hdd. I sold the machine last oct 2011 with a year on applecare. So this year all 2009 iMacs with seagate hdds were recalled due to the hdd going bad. Apple lost this bet because the hdd's failed too soon. They ended up recalling all the 2009 iMacs. Why? no 2 screw access panel in the back of the machine to replace the hdd.

Look I get that making a machine that lasts too long is bad for business.


Here is a great tool a shun knife from the MH series.


http://www.amazon.com/Shun-Steel-9-...e=UTF8&qid=1353365434&sr=8-6&keywords=shun+mh

No longer sold why?
The handle does not break down this set of shun knives ( mh for metal handle) was just about the most indestructible hi end knife series ever made. Shun stopped production because they were too durable.

These knives are dishwasher machine washer safe! This is not the case with hi end knifes in general as the handles do poorly in dishwashers as does most really good steel.

I upgrade and sell 80 to 120 macs a year. I will push the dual to most users for 2012. I won't sell the quads to users that don't understand computers that well. If I sell a quad to someone that really wants one I will only sell it with a ram upgrade. I do not want warranty issues..


I will put ssd's in the dual base model with no warranty fears. (If the gear has an issue I would sell warranty a dual for a buyer but not the quad.)
 

motrek

macrumors 68030
Sep 14, 2012
2,613
305
all this reminds me of the 2009 27 inch iMacs .. I complained I bi-atched about the encased hdd. I sold the machine last oct 2011 with a year on applecare. So this year all 2009 iMacs with seagate hdds were recalled due to the hdd going bad. Apple lost this bet because the hdd's failed too soon. ...

Why are you talking about iMacs and Seagate hard drives when I asked you to substantiate your claims about Mac Minis? Absolutely irrelevant.

Besides, the iMac thing doesn't prove anything. Apple and many other vendors got unlucky with those Seagate hard drives. It doesn't say anything about Apple or the design of the iMac. Yes, Apple has to eat the cost of their "genius bar" people replacing the drives but it only takes a few minutes of work... you just use suction cups to get the glass off, unscrew the monitor (9 screws I think), and then you have access to the drive. Apple isn't the only company that has to occasionally recall its products for whatever reason. Every company has done millions of laptop battery recalls. Dell did major motherboard recalls due to faulty capacitors, etc. Car manufacturers are constantly doing recalls. etc. It's not like you're a genius for predicting that Apple might have to recall some product at some time for some reason.
 

philipma1957

macrumors 603
Original poster
Apr 13, 2010
6,367
251
Howell, New Jersey
Why are you talking about iMacs and Seagate hard drives when I asked you to substantiate your claims about Mac Minis? Absolutely irrelevant.

Besides, the iMac thing doesn't prove anything. Apple and many other vendors got unlucky with those Seagate hard drives. It doesn't say anything about Apple or the design of the iMac. Yes, Apple has to eat the cost of their "genius bar" people replacing the drives but it only takes a few minutes of work... you just use suction cups to get the glass off, unscrew the monitor (9 screws I think), and then you have access to the drive. Apple isn't the only company that has to occasionally recall its products for whatever reason. Every company has done millions of laptop battery recalls. Dell did major motherboard recalls due to faulty capacitors, etc. Car manufacturers are constantly doing recalls. etc. It's not like you're a genius for predicting that Apple might have to recall some product at some time for some reason.

I am not a genius. My IQ is too low 149 or 151 on the two iQ tests taken years ago not 160. Those scores would have ranked me at about 1 in 1000. Since it drops off a bit with age I most likely dropped to 140 or so which makes me 1 in 135 or so. Which means in the usa alone 2 or 3 million people have higher iqs then I do. World wide more like 50 million are higher then me.

So for brains I not dumb, but really smart (ie Genius) no that is not me.


Plus my claims are on quad core mac mini mobos from 2011 and 2012. Not mac minis. Just a particular subset of minis. Also even if you own that particular machine you would need be a certain type of user to have to worry.

I will be able to substantiate them( my bad mobo claims FOR QUAD CORE MINIS ) around 2014 for the 2011's and around 2015 for the 2012's.

See your thinking on the 2009 iMacs is exactly what I said ;

Apple got unlucky on the drives they did not last 3 plus years. Their bad luck was a residue of their design. ie no back door for the hdd. They gambled and lost on that design.

My predication is this is likely on the 2011 quad core minis and the 2012 quad core minis . Of course this would be more expensive error on their part. Putting in a new mobo in a 2011 mini or 2012 means a new cpu since it is more or less permanently attached.

so If I sell quads this year i will try to sell with 3yr plans not 1 yr.


If I recall your post earlier you say (guess?) that apple engineers tested this with long term cpu stress tests. 'weeks' is your number.


well a 2 or 3 or 4 weeks stress test of a quad core cpu is not much if you want the machine to run 24/7 365 as a server.

I know a few small server farms that wanted mini quads last year. low power use and small size are keys for them.

They have all the quad minis side mounted with extra cooling. As a precaution to keep the quads temps under 100c.
 
Last edited:

motrek

macrumors 68030
Sep 14, 2012
2,613
305
...
I will be able to substantiate them( my bad mobo claims FOR QUAD CORE MINIS ) around 2014 for the 2011's and around 2015 for the 2012's.
...

You said "based on 2011 server returns ( high rate of bad mobos) you are not correct" as a statement of fact, i.e., you KNOW that RIGHT NOW (not 2014) there has been a high rate of bad motherboards resulting in a high rate of returned product.

So you should either be able to substantiate that claim right now (not 2014) or you should phrase things differently, like, "a lot of 2011 Mac Mini motherboards will go bad in the 2014 timeframe based on nothing more than a poorly informed hunch."
 

propower

macrumors 6502a
Jul 23, 2010
731
126
The quad mini mobo certainly has the largest stress of the Mini family (and has only been around since 2011). It does make one curious about the quad MBP mobos as well (run very much in the same way!). Still kind of sad to me but it is design choice they are making.

I think I will stick by my use the Mini, wait to see what the new MP looks like and either go that way or get the 3 yr applecare and make sure to sell before it runs out :).

A bit of a strange new paradigm for computers and me! I don't push my machines really hard but they don't sit at 5% all day either!
 

Darby67

macrumors 6502
Qq

C'mon Philip, you are always making these sorts of claims (Minis are designed to fail in 3 years, 2011 servers all got returned, etc.) which you would have no way of knowing unless you worked at Apple.

Can you substantiate anything you've been saying lately?
Yes, i do. But i am afraid that's all i can tell. I signed a NDA.


I am not a genius. My IQ is too low 149 or 151 on the two iQ tests taken years ago not 160. Those scores would have ranked me at about 1 in 1000. Since it drops off a bit with age I most likely dropped to 140 or so which makes me 1 in 135 or so. Which means in the usa alone 2 or 3 million people have higher iqs then I do. World wide more like 50 million are higher then me.

So for brains I not dumb, but really smart (ie Genius) no that is not me.

Both these statements are comical. Your dislike for what Apple has done comes down to basing your case on your mysterious dream CV and chest thumping IQ deflection.
 

motrek

macrumors 68030
Sep 14, 2012
2,613
305
The quad mini mobo certainly has the largest stress of the Mini family (and has only been around since 2011). It does make one curious about the quad MBP mobos as well (run very much in the same way!). Still kind of sad to me but it is design choice they are making.

I think I will stick by my use the Mini, wait to see what the new MP looks like and either go that way or get the 3 yr applecare and make sure to sell before it runs out :).

A bit of a strange new paradigm for computers and me! I don't push my machines really hard but they don't sit at 5% all day either!

It is a brave new world, since these new Intel chips will happily overclock themselves until they hit ~100C, which is a very deliberate design decision on Intel's part, and you KNOW they've done their due diligence to make sure these chips will work for many many years at that temperature.

From reading various articles online, many PC laptops experience throttling (i.e., less than maximum Turbo Boost) meaning that their CPUs are running at ~100C, so Apple's designs don't seem to be any better or worse than other vendors' designs. (Actually maybe better, because I've run some tests on a friend's MacBook Air and the CPU did get up to the mid-90s but it never throttled.)

If it turns out that a 100C CPU causes undue stress on a motherboard then everybody who's bought a laptop in the last couple years is in for a world of hurt... but the engineer in me thinks that modern motherboards must be designed to withstand these kinds of chip temperatures.

The only case I'm aware of with chip temperatures being an issue for motherboards is with the XBox 360 and the red ring of death issue. I think that ended up being a problem with the solder formulation more than chip temperatures.

Thinking about it, I recall that it's fairly common for the PWM chips on motherboards to run well over 100C and that temp will fluctuate with CPU load, so maybe it's not really a big deal at all for motherboard durability for the CPU to run at these temperatures.
 

Snowcake

macrumors regular
May 18, 2010
187
0
Image
So if i understand it correctly, the strange behavior of cputest is the bug, not the actual "failed tests"?
In the picture above you see a section which displays "Failed tests"
The number of that was always 0 during the bug? It has never showed a another number than 0 on your second/replacement Quad Mac Mini? Correctly said, it had no errors but your problem was something else? (A bug)

Philipma1957 is probably ignoring my question because CPUTest did show errors. (I had asked it twice) :)
 

MJL

macrumors 6502a
Jun 25, 2011
845
1
I know that there are electronics that are designed for high temperatures (that we were only dreaming off not that long ago) for special applications but they are not packed with your run-of-the-mill parts which obviously Apple is using.

Perhaps those under you that think that Apple engineers have designed a proper machine should look up the ratings for the CR2032 CMOS battery - it is designed for 60C max. With extended running at 100% CPU (and throttling) it does reach that temperature.

http://en.wikipedia.org/wiki/CR2032_battery

There are also already a number of the 2.5 Ghz (mid model) Mac mini 2011 that have had their mother board replaced. Hear those that are bitching that it happened when their computer was just out of the 1 year warranty.

The Ivy bridge CPU is supposedly running cooler than the Sandy bridge under "normal loads" and only running hotter when being "over clocked" yet Apple has modified the fan. To me it is an overt acknowledgement that there is an issue.

For me being a gambling man - I am not taking the risk that my computer is crapping out at a crucial moment. Do not misunderstand me - I love the small form factor with build in powersupply - easy to travel with.

----------

Not sure why you think 7 years is a big achievement.

3 plus 7 makes 10 in my calculation.

And before having Apple computers I NEVER had the need to have a motherboard replaced. CD/DVD players, screens and HDD's, yes, but no motherboard failures.
 
Last edited:

propower

macrumors 6502a
Jul 23, 2010
731
126
I have no answers for what the current design margins in Apple computers are. From an old designers viewpoint I can say that 100degC is a tricky place for electronics. REally above the commercial range of parts (85degC) and a little below the full Mil-Spec range 125degC. But have the following thoughts anyway...

There are many formulations of PWB material that can do high temperature with no problem. Some of the really nice stuff is 180degC rated. What is Apple using? I really don't know. In most commercial stuff, FR4 is the common one and 130degC rating is usual. But in the hi-rel world it is all about derating margin.

125degC rated IC's (PWM, CPU, analog etc) are easy to find but are WAY more expensive than there commercial counterparts. Is Apple using choice parts in critical areas?
 

motrek

macrumors 68030
Sep 14, 2012
2,613
305
...
There are also already a number of the 2.5 Ghz (mid model) Mac mini 2011 that have had their mother board replaced. Hear those that are bitching that it happened when their computer was just out of the 1 year warranty.

The Ivy bridge CPU is supposedly running cooler than the Sandy bridge under "normal loads" and only running hotter when being "over clocked" yet Apple has modified the fan. To me it is an overt acknowledgement that there is an issue.
...

First of all, any computer component can fail for any number of reasons. I've done a few searches today for 2011 Minis breaking for any reason and wasn't really able to come up with anything. Apparently you saw a couple reports of motherboards being replaced. I can't find them and if this was a design flaw that happened to everybody who stressed their Mini then there would be an Apple Support Forum thread a mile long along with a possible class action lawsuit, since a product shouldn't break when used as intended and lawsuits get filed for this sort of thing all the time.

Second, Apple made a highly publicized change to their fans recently to make the spacing between the fins uneven because that smoothed out the noise character. In other words, the change to the fan in the Mini was only to make the computer sound nicer and isn't an admission of anything.

----------

...
125degC rated IC's (PWM, CPU, analog etc) are easy to find but are WAY more expensive than there commercial counterparts. Is Apple using choice parts in critical areas?

Eh, Intel is very obviously designing their CPUs to reach 100C under very pedestrian operating conditions so I have to assume it's okay.
 

MJL

macrumors 6502a
Jun 25, 2011
845
1
First of all, any computer component can fail for any number of reasons. I've done a few searches today for 2011 Minis breaking for any reason and wasn't really able to come up with anything. .

two things:

- you did not address teh 60 degree temperature
- you cannot find the details with google, how I came to this knowledge I cannot disclose (unfortunately)

and since you seem to think Apple cannot anything wrong you might want to think about this:
http://www.seattlerex.com/seattle-rex-vs-apple-the-verdict-is-in/comment-page-1/
or in case of the macbook pro this:
http://www.patentlyapple.com/patent...apple-for-defective-macbook-logic-boards.html

Lenovo had a similar issue with a GPU on the early Thinkpad T61P and they went all out to find the error and once discovered there was a free replacement. No legal wrangling and bullying going on. (The fault primarily occured when people started to use that laptop for gaming and were stressing the computer.)

ah well, that is more than two things. Happy reading and I look forward to the day that Apple is forced of its pedestal (just like Microsoft, what was their highest share price and what is it now? Never mind)
 

motrek

macrumors 68030
Sep 14, 2012
2,613
305
two things:

- you did not address teh 60 degree temperature
- you cannot find the details with google, how I came to this knowledge I cannot disclose (unfortunately)

and since you seem to think Apple cannot anything wrong you might want to think about this:
http://www.seattlerex.com/seattle-rex-vs-apple-the-verdict-is-in/comment-page-1/
or in case of the macbook pro this:
http://www.patentlyapple.com/patent...apple-for-defective-macbook-logic-boards.html

Lenovo had a similar issue with a GPU on the early Thinkpad T61P and they went all out to find the error and once discovered there was a free replacement. No legal wrangling and bullying going on. (The fault primarily occured when people started to use that laptop for gaming and were stressing the computer.)

ah well, that is more than two things. Happy reading and I look forward to the day that Apple is forced of its pedestal (just like Microsoft, what was their highest share price and what is it now? Never mind)

I think you misunderstand me. I like Apple products in general but I'm not an Apple rep. I don't have to respond to every point you make in order to paint Apple in a rosy light and I don't have to deny the reality of Nvidia recalls, or Seagate recalls, or whatever recall you want to bring up.

In fact, from the links you bring up, I've probably been using Apple products off-and-on more than you have and I have been personally affected by their various design flaws. I had a Mac laptop back in ~97 which had a battery and touchpad recall. I used G3 towers in college with the faulty power buttons that would (very) occasionally catch fire, and the G4 cube with the cracked plastic. I've owned two MacBooks with chipped top cases.

What's annoying me in this forum, and this thread, though, is that people are just making s*** up, completely unsubstantiated. I have an honest interest in knowing how the Mac Mini behaves under load (how loud it is, in particular), and I come here and people are just telling fairy tales about 2011 Mini motherboard failure rates with literally zero evidence other than their crazy-old-man theories, and waving their hands about parts overheating when, for all anybody knows, the parts are absolutely engineered to operate indefinitely at these temperatures.

So now you're the 2nd person in this thread claiming extra-special insider information about mobo failure/return rates. Come back to me when you can point to a big thread on the Apple Support Forums. For crissake, I can find a million threads and posts about HDMI-out not working right on the new Mini but I can't find a single thread with somebody claiming an overheated/broken 2011 Mini, yet you guys are acting like it happens all the time. What's wrong with this picture?
 

dasx

macrumors 65816
Jun 18, 2012
1,107
18
Barcelona
(Actually maybe better, because I've run some tests on a friend's MacBook Air and the CPU did get up to the mid-90s but it never throttled.)

Out of curiosity. How do you check that? I'd like to see if my Mini does throttle at some point. :)
 

philipma1957

macrumors 603
Original poster
Apr 13, 2010
6,367
251
Howell, New Jersey
Philipma1957 is probably ignoring my question because CPUTest did show errors. (I had asked it twice) :)

the cpu test never showed a number other then 0. for errors.


it just stopped working. if it was the quad core with 8 cores at 100% clicking down the test count. it would go to 1 core at 100% and the test count stopped clicking down.

----------

the cpu test never showed a number other then 0. for errors.


it just stopped working. if it was the quad core with 8 cores at 100% clicking down the test count. it would go to 1 core at 100% and the test count stopped clicking down.


motrek this is a quote from you above;

"So now you're the 2nd person in this thread claiming extra-special insider information about mobo failure/return rates. Come back to me when you can point to a big thread on the Apple Support Forums. For crissake, I can find a million threads and posts about HDMI-out not working right on the new Mini but I can't find a single thread with somebody claiming an overheated/broken 2011 Mini, yet you guys are acting like it happens all the time. What's wrong with this picture? "

you are selectively reading your own interpretation into my thread. that is what is wrong with your attack.





the bottom line is that quad users that run the machine 24 /7 / 365 with high cpu use will need extra cooling. Most Users on this forum need not worry.
 
Last edited:

motrek

macrumors 68030
Sep 14, 2012
2,613
305
Out of curiosity. How do you check that? I'd like to see if my Mini does throttle at some point. :)

Some of the software I write goes through a bunch of iterations and prints out how long it takes to do each iteration.

At the end of the run, the CPU was in the mid 90s but still chugging along just as fast as when it started.

I plan on buying the base i7 Mini on Black Friday this week and will be able to see if it throttles or not then.
 

philipma1957

macrumors 603
Original poster
Apr 13, 2010
6,367
251
Howell, New Jersey
Some of the software I write goes through a bunch of iterations and prints out how long it takes to do each iteration.

At the end of the run, the CPU was in the mid 90s but still chugging along just as fast as when it started.

I plan on buying the base i7 Mini on Black Friday this week and will be able to see if it throttles or not then.

the base runs 20 to 30 f cooler it should be fine.

I was unable to get it past 190f and cpu use stayed at 99 to 100%. I like the base mini a lot. I run it with an external t-bolt ssd no muss no fuss.
 
Register on MacRumors! This sidebar will go away, and you'll see fewer ads.