Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
That's not actually true though.

The 970 EVO plus NVMe is about half as fast as what I've got in my MBP. I use one with a backup disk connected via a thunderbolt bridge.

The 980 PRO is closer but still no banana.
If you have thunderbolt 3, then you are bottlenecks by the interface and not the disk.

The 980 pro is just as fast as the M1 max
 
Well are you sure about that bub?
lots of unknowns on that analysis of the SSDs. it's even pointed out in the article.

from my personal experience in the 2014 to 2018 timeframe, we had nearly 50% failure rate of our boot SSDs (10 of 21). the failures were all from the same manufacturer and
more telling, the same batch (nearly sequential serial numbers). the others are still chugging along (different manufacturer)
 
Yep. We just need the latency to improve. Then we can finally remove RAM and just use our SSD as RAM as instant access.

Imagine running your GPU with 1TB of memory without the need to transfer data

Or what about no more HDDs and SSDs, just computers running on direct CPU-registers and secure non-volatile hyper-capacity hyper-fast RAM (terabytes). Only non-bloated non-SaaS open source operating systems taking only kilobytes, maybe megabytes in size. Apps are either free forever or one-time-only (non-subscription) based. Apple's, Google's, and Microsoft's extremely huge and bloated force-updating bloatware spyware operating systems would be irrelevant and useless for the world. Instead we could use powerful privacy-respecting people-friendly and energyefficient hardware with featherlight non-bloated software. An Internet where privacy is default.

There are already some companies slowly but surely working on this, like Purism with their Librem 5 phone, and Proton Technologies AG with their ProtonMail. In about 10 years or less, huge companies like Microsoft is no longer relevant. Yes, sorry for being slightly off-topic, I'm just trying to show that software companies depends on hardware companies and vice versa. Anyone remember Nokia with their Nokia 3310 phone released in 2000 (over 2 decades ago)? A GSM-only phone with only 1 kilobyte of memory, but look at all the things it could do.
 
lots of unknowns on that analysis of the SSDs. it's even pointed out in the article.

from my personal experience in the 2014 to 2018 timeframe, we had nearly 50% failure rate of our boot SSDs (10 of 21). the failures were all from the same manufacturer and
more telling, the same batch (nearly sequential serial numbers). the others are still chugging along (different manufacturer)
And someone will always be part of the statistics. My MacBook Had it’s GPU controller fail making my computer a brick, fortunately I could salvage the Data by removing the storage before I sent it on repair

The bigger problem is generally not the ssd failing but the motherboard or other component of the computer failing that will make the SSD unsalvageable. But if you do have the SSD fail you will be left with:
Expensive repair bill for essentially a cheap fix
Or
An expensive paperweight
 
People seem to ignore the fact that not only Apple customers need SSD flash, nor any Apple device can be purposed similarly to a generic USB drive (ignoring the price).

Apple will definitely raise the price of their computers soon to cover this incident while maintaining or even raising their profit margin thanks to this, instead of the other way around.

Now I really need to grab that 2TB SSD I was eyeing on for a long time before the price double or even triple. Housing 500GB iTunes library on internal SSD isn’t feeling great.
 
I would dare to say it’s one of the most important part. Everything else you can replace or repair. But if the storage gets ****ed, then your data is forever lost.

You could even do an instant transfer to a new computer in the case of catastrophic failure IF it wasn’t soldered.

At least apple used to have a dedicated port on the motherboard you could use for data recovery. Now it’s impossible
SSD controllers give us a nice countdown though. Better a numeric parachute than a sudden clunk. In fact our enterprise SSDs at work in DB servers get replaced every year or so towards the end of their life. As for every other computer in an organisation of 500 people, we haven’t had a single SSD failure on a laptop or desktop yet. we’ve seen some crazy high TBW on some dev computers as well.

My entire computer is backed up weekly onto an external disc and constantly mirrored to iCloud anyway so the opportunity for damage is low. We should all bear that in mind. Recovery planning should be done ahead of time ?
 
Or what about no more HDDs and SSDs, just computers running on direct CPU-registers and secure non-volatile hyper-capacity hyper-fast RAM (terabytes). Only non-bloated non-SaaS open source operating systems taking only kilobytes, maybe megabytes in size. Apps are either free forever or one-time-only (non-subscription) based. Apple's, Google's, and Microsoft's extremely huge and bloated force-updating bloatware spyware operating systems would be irrelevant and useless for the world. Instead we could use powerful privacy-respecting people-friendly and energyefficient hardware with featherlight non-bloated software. An Internet where privacy is default.

There are already some companies slowly but surely working on this, like Purism with their Librem 5 phone, and Proton Technologies AG with their ProtonMail. In about 10 years or less, huge companies like Microsoft is no longer relevant. Yes, sorry for being slightly off-topic, I'm just trying to show that software companies depends on hardware companies and vice versa. Anyone remember Nokia with their Nokia 3310 phone released in 2000 (over 2 decades ago)? A GSM-only phone with only 1 kilobyte of memory, but look at all the things it could do.
I think you underestimate the amount of code it takes to do stuff and the amount of effort it takes to produce that code and the amount of engineering it takes to throw a CPU on the market.
 
SSD controllers give us a nice countdown though. Better a numeric parachute than a sudden clunk. In fact our enterprise SSDs at work in DB servers get replaced every year or so towards the end of their life. As for every other computer in an organisation of 500 people, we haven’t had a single SSD failure on a laptop or desktop yet. we’ve seen some crazy high TBW on some dev computers as well.

My entire computer is backed up weekly onto an external disc and constantly mirrored to iCloud anyway so the opportunity for damage is low. We should all bear that in mind. Recovery planning should be done ahead of time ?
Haha true that but it would really Suck to have that coffee spill in the middle of the week.
 
Haha true that but it would really Suck to have that coffee spill in the middle of the week.
I actually did that to my 2010 MBP and didn’t have any cloud backup software on it. I was down for 3 days in the middle of a contract and lost 3 days work. Now it’d be 30 minutes of swearing, a 30 minute drive to my local apple store (who have a bunch of 14” MBPs in stock) and just grab another machine.

Granted it’d financially burn but I spent the last decade making sure I only buy stuff I can afford to replace.

Edit: I also don’t eat and drink near my computers now ?
 
  • Haha
Reactions: Sophisticatednut
Well are you sure about that bub?
Unless there's a spread of different types of SSD this is more anecdotal than fact. If you dig into Blackblaze's data you can see there's certain HDD's that skew the result heavily. I'd need to see a similar breakdown for SSD's. It could be one SSD brand has issues whilst others perform reliably.

Also I was going off my own anecdotal experience with SSD's and since buying my first OCZ 64GB SSD over a decade ago I have never had a failure or problem.
 
  • Like
Reactions: lcubed
how times have changed, a while back it would have been China making these mistakes, now korea and china have overtaken Japan
 
Unless there's a spread of different types of SSD this is more anecdotal than fact. If you dig into Blackblaze's data you can see there's certain HDD's that skew the result heavily. I'd need to see a similar breakdown for SSD's. It could be one SSD brand has issues whilst others perform reliably.

Also I was going off my own anecdotal experience with SSD's and since buying my first OCZ 64GB SSD over a decade ago I have never had a failure or problem.
Well we do know they break, and someone will always be that unlucky bastard. I just think it’s kind of wasteful if the computer becomes expensive paperweight if the storage dies or that storage gets impossible to access if something else breaks ?

And I would say server hardware might be a little bit more resilient than the stuff consumer generally use, as the lowest quality
 
Does anyone know what kind of contamination are they talking about? In every other place including their press release it's just "unspecified contamination" or "contamination of certain material used in its manufacturing processes".
Jonny (John for short) from HR dropped his cup of Starbucks latte in the large chips mixing container. He was fired.
 
Apple used other suppliers forSSD besides WD. Even Samsung supplies Apple with SSD
As for the price most likely will go up but hopefully not drastically
 
I feel so sorry for people that believe that everything is a conspiracy. Heads may roll in the QC department depending on the reason that this escaped inspection.

As noted by others WD is a publicly traded company. This contamination problem will likely result in a write-off/write-down of hundreds of millions of billions of dollars. That is something the company is obligated to inform their shareholders of.
Companies for years have been manipulating prices of products and cartel-like behaviour and it's nothing new, no matter if public or private. The fact remains the term "contaminated" could mean anything and seems like a way too simple one word explanation of the issue. Im intrigued if this was done deliberately by the company / staff to manipulate pricing.

From my knowledge most of the inspections of NAND are done by robotics, not humans. Claim a "contamination" and pricing rises, so in effect they are profiteering from the issue. Its very convenient this occurred just as NAND prices were dropping. It's not for the consumer to pick up the financial slack when they caused the problem as WD is entirely liable and costs should not increase because of this.
 
  • Like
Reactions: nebojsak
The fact that Apple's markup is outrageous is compounded by the fact its not replaceable or upgradable. The difference in getting something like a Lenovo is that your not stuck with the limited choices they offer you. I can put 2 off the shelf NVME drives and 2 dimms in something like a Legion 5. If they are using WD NAND, its most likely equivalent to the stuff in the Gen 4.0 WB Black SN850 which 2TB is $250. Those drives have 5 year warranties.

The catch is that Apple is really charging you for the privilege of longevity, because wear leveling increases with size. That is why there isn't really any performance scaling between the M1 Air and M1 MBP. If you did intensive work on the base M1 MBA, the 8GB will result in more paging to the SSD, and the 256GB will die sooner than all higher capacity models. The vast majority of failed SSDs I see in (doing computer repair) PC and Mac have been 256GB and lower, with very few 512GB and above. I personally have a 8+ year old 512 GB SSD with smart errors but still works. It's outlived the usefulness of several devices, but its wonderful that I could keep using it in multiple devices. Apple when through like 6 proprietary removable SSD iterations, when standard M.2 NVME would have been perfect. Heck I see adapters in some people's macbooks that let them use standard NVME drives. Their designs continually influence unnecessary waste.
 
  • Like
Reactions: mi7chy
Register on MacRumors! This sidebar will go away, and you'll see fewer ads.