No you have totally misunderstood how cpus are made classified and work. You think you know what you are talking about I suggest you really don’t not that I care mind you just pointing out nonsense when it’s written. Yes there are transistors In a cpu where if they go your cpu is dead but certainly not all of them or binning would be impossible.
Got a tip for us?
Let us know
Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
No you have totally misunderstood how cpus are made classified and work. You think you know what you are talking about I suggest you really don’t not that I care mind you just pointing out nonsense when it’s written. Yes there are transistors In a cpu where if they go your cpu is dead but certainly not all of them or binning would be impossible.
I think you are talking about two very different things. Toutou was replying to the following post by Spytap
While I am not a semiconductor expert by any definition of the word, I think its a bit more complicated and I believe that both posters make good points in their own ways. To sum it up for the impatient (detailed reasoning follows): I don't believe that ageing will directly slow down the CPU, however it will probably increase the heat/energy consumption, which will might slow the CPU down indirectly.
First of all, Spytap is definitely right that semiconductors deteriorate with time, and yes, they do become slower (need more energy and/or time to change states). However, I think that Toutou is also correct: CPUs are discrete digital devices that run on a clock. So their performance is conditioned on a clock signal generator and not on the maximal performance that the silicon is actually able to run at— that will be different for various parts of the CPU anyway due to imprecisions during manufacturing. So gradual slowing down of the circuitry as the damage through age is accumulated will not directly affect the performance, as long as the switches are still fast enough to work with the common clock. And AFAIK the clock is on much lower precision that what a transistor can put up (if it weren't the chip would simply not work), so these slowdowns are negligible. And if at some point parts of the circuitry are indeed too slow, your processing will simply break down and you will suffer a permanent failure, since the CPU simply can't work anymore in the way it was designed. So gradual slowing down will not make the CPU slower, it will simply make it fail at some (potentially very remote) point in time. It is possible that the clock generator will also slow down with age, but I am not aware of any research that suggests that these effects are measurable.
At the same time, it is entirely possible that ageing will increase the energy requirement for the circuitry to work as specified. So the CPU would output more heat to maintain the same performance levels and will not be able to reach the same performance peaks as before due to throttling/power system limitations. This might be the effect Spytap is talking about.
Binning is a different thing that has nothing to do with ageing. Its simply that — as mentioned above — manufacturing is subject to random imprecisions and other problems, so that different exemplars of the same chip (on paper) end up being of different quality. Some can't clock as high/need more energy and some are simply broken. Binning is a way to split the chips into performance/quality groups so that manufactures can still sell most of the batch and not throw perfectly usable (even if not as "good") chips away. Note: I am sure that you guys know all this, just writing it up so that readers who are not aware of these things have some context.
Edit: here is a very insightful discussion about all the relevant phenomena — https://www.reddit.com/r/askscience...worn_and_loose_performance_gradually/cmhwm9e/
Last edited:
No they don’t, only if you opted to buying an entry level base model MBP, if you went with the highest spec, usually the “late” generation of each year after the 2nd refresh / updated CPU/GPU, then you can expect to have steady and maintained quality of use. I’ve used my highest spec mid 2006 17”MBP till 2011 when I got the late 2011 top spec’d 17”MBP, Then I had the late 2013 top spec’d 15” rMBP, I sold all of them and afforded the late top of the top spec’d 2017 15” MBP w/T.B. Non of the above slowed down at all except the 2006 developed a strange “constant on roaring fan” issue, ,that was about it, it never lagged or crashed ever.
I had two MBP's develop issues around the 3-year mark. There is definitely a price to be paid for making products too thin with regards to long-term reliability. Never seem to get the same issues with old Thinkpads or even old Latitudes for that matter.
You have been exceptionally lucky then. Most Lenovos and Dells I had the displeasure in dealing with were broken after only two years.
I haven't owned a Lenovo for years (like 2008) and just recently ordered a X1 Carbon (6th gen) as my MBP 2017 had some keyboard issues. I have a Dell XPS 13 from 2016 and it's held up so far. Was the Lenovo lemon you had a recent manufacture?
Depends which ones you buy. If you buy the cheap consumer-level stuff then those don't last too long. The business class stuff like thinkpad's and latitudes tend to be built quite well.