Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
Was going say I was pretty sure I've seen him post on Macrumors before and ... <looks up> yep. :/
Yea. I mean, if people are literally trying to accuse me of being an alt account for Rossman, I guess I must be coming across more articulately than I thought (do people really think I'm THAT significant? 😂) I have a feeling they're just trying to get under my skin, but hey, that's just an assumption. I can't read anyone's mind, I'll give them the benefit of the doubt.

All is well that ends well I suppose. I actually kinda take it as a compliment, although rest assured, I'm most certainly not Rossman, nor do I have any association with him on any personal or professional level.
 
Last edited:
  • Like
Reactions: BigMcGuire
Big deal. Buy AppleCare and then Apple pay to have it fixed.

The days of user replaceable SSD and RAM are OVER. The real reason is performance. The new macs are using the SSDs more like RAID where the data is distributed over multiple chips. Using two is twice as fast as using one.

Also, by soldering, they can make the computer smaller and remove a common failure pint, the connector.

Yes, I know it was fun to buy a low-spec computer, then hunt the Internet for bargain parts and upgrade at low cost. But you can't match the reliability and performance of the new design.

The idea that devices must have soldered and highly wearable components such as SSDs for thinness is such a braindead argument.

Laptops which are just as thin or thinner than current Macs have hot-swappable NVMe SSDs.
 
! Haven't watched it yet.


Edit: Pretty cool, he calmly reads through quite a few of the comments and explains quite a bit.

@l.a.rossmann Thank you for your videos, definitely learned something, as I have on quite a few of your other videos.
I wasted few minutes watching. He is bigger clown than I thought. Throw some random crap about same argument, get more clicks and make more money. Out of here, not gonna bother anymore.
 
  • Haha
Reactions: 120FPS
I wasted few minutes watching. He is bigger clown than I thought. Throw some random crap about same argument, get more clicks and make more money. Out of here, not gonna bother anymore.
You're definitely entitled to your opinion. I watched the whole thing and found I learned a lot about quite a few of the issues people were talking about in this thread - he takes the time to explain the background on quite a few of the topics brought up - and obviously provided his opinion. Thought that was pretty cool.
 
For the Mac's just two losses both Air's one no thx to a glass of water and another seemingly random. It's not a bad record given the time frame of well over two decades...

I've had a 2020 MacBook Air survive about 100ml of espresso direct to the keyboard while turned on. I turned it upside-down to drain it out. It's still working today.

And backups are critical irrespective of hardware reliability. Theft happens. If you don't have reliable backups, rectify that asap - all hardware dies and all of it is prone to theft or destruction.
 
  • Like
Reactions: ArkSingularity
You're definitely entitled to your opinion. I watched the whole thing and found I learned a lot about quite a few of the issues people were talking about in this thread - he takes the time to explain the background on quite a few of the topics brought up - and obviously provided his opinion. Thought that was pretty cool.
It’s not an opinion though. I doubt he held any real job in software/product/project development. I started laughing when he butchered the word defect. I sent the link to couple of corporate product/program management trainer buddies of mine. It’s a defect when apple says product does a, b, and c but delivers x,y, and Z. Lot of young professionals will learn what is not a defect, what is a design constraint/trade off. Thanks to his very educative video.
 
  • Like
Reactions: Analog Kid
It’s not an opinion though. I doubt he held any real job in software/product/project development. I started laughing when he butchered the word defect. I sent the link to couple of corporate product/program management trainer buddies of mine. It’s a defect when apple says product does a, b, and c but delivers x,y, and Z. Lot of young professionals will learn what is not a defect, what is a design constraint/trade off. Thanks to his very educative video.
Does it matter if he has experience in software development? He's speaking as an experienced repair professional from a hardware engineering standpoint. His business (which you can Google) has existed for years and has professionally repaired Apple products, including some products that Apple themselves was unable (or declined) to repair. Actually, he has been doing repairs longer than he has been doing youtube, and repairs/hardware is the subject of this conversation. Software engineering experience is irrelevant. (That's like saying a dentist is unqualified to diagnose your teeth because they don't know how to do open heart surgery.)

(Side note: I also find it interesting that a lot of the folks who are accusing me, and by extension, folks who may see some of Rossman's side on this, of being conspiracy theorists while simultaneously forgetting to even check my profile before accusing me of being an alt account for Louis Rossman. I'm not gonna make assumptions about this, I just find it interesting. 😂)
 
Last edited:
Better question is, when is the last time someones SSD died? Outside of anomalies most will last longer than the life of the machine especially in Apple’s case when it can’t be moved from computer to computer.

I don’t mind Louis but he obviously has an agenda and as someone who grew up building pc’s I just don’t care anymore about swapping parts but I do expect them to last. If I felt strongly about it I simply wouldn’t buy it.
I had this issue that Louis talks of which is why I made this post.

 
Does it matter if he has experience in software development? He's speaking as an experienced repair professional from a hardware engineering standpoint. His business (which you can Google) has existed for years and has professionally repaired Apple products, including some products that Apple themselves was unable (or declined) to repair. Actually, he has been doing repairs longer than he has been doing youtube, and repairs/hardware is the subject of this conversation. Software engineering experience is irrelevant. (That's like saying a dentist is unqualified to diagnose your teeth because they don't know how to do open heart surgery.)

(Side note: I also find it interesting that a lot of the folks who are accusing me, and by extension, folks who may see some of Rossman's side on this, of being conspiracy theorists while simultaneously forgetting to even check my profile before accusing me of being an alt account for Louis Rossman. I'm not gonna make assumptions about this, I just find it interesting. 😂)
It does, when he ventured in to design defects, and developement choices. The responses to his videos are strong because of clickbait/sensational/trollish slant. I said it earlier Apple has poor repairability, i havent bought an apple desktop/workstation in over a decade, because i value ability to upgrade or repair in my workstations. If he had the video with "I give Apple 0/10 for repairability for choosing security encryption and other factors over repairability". I woundt have problems, if not for half truths.

Its funny, he throws around word like defects, try keeping a job in product delivery/developement with that understanding of defect. No company will deliver a product. lol.
 
This seems to be making assumptions about his intentions much more than it is about addressing his actual arguments, but I'll leave it at that.

You shared a report showing how much money he likely makes from YouTube and asked me to clarify my view on it, which I did.

If you want me to address Rossmann's arguments in that video, I can't because I won't watch it. He lacks a basic engineering sense and his videos I've seen linked here in the past have been demonstrative nonsense. I won't waste my time or credit him any additional views.



If you want my views on soldered flash chips, I think I've made myself clear but here it is again:

Sockets are a common failure point. They have moving parts. They are assembled by fat fingered humans, not micron accurate machines. The high speed electrical connection is made by pressure, not metallurgy. Corrosion can happen at the interface. They wear under mechanical stress and vibration. If you want a reliable electrical connection solder it, don't push it together.

Touching things breaks them. Every time you open something, touch something, put your finger oils or accumulated static charge on something, you degrade its reliability. Every time you manipulate a socket or connector you wear it and reduce its reliability. Pins bend. Opening things violates seals, if not reassembled correctly you can stress things, breach EMI shielding, breach moisture sealing, fail to ground things properly, loosen screws leading to mechanical wear over time, over tighten screws leading to stress and fatigue, forget to properly power things down, bridge live electrical lines damaging components even if you have powered it down, short a battery pack, etc, etc, etc. If you want to make something reliable, don't touch it and don't make it easy for other people to touch it.

These types of failures are rather insidious and do not show up immediately. They can take months to manifest. Most often a first failure might be a system failure but a second failure is because of the human intervention.

When a part fails, it's very often not the source of that failure. It could indicate a failure elsewhere in the system where replacing a module will just lead to another future failure in that part or another because the underlying problem was not resolved by replacing the module. Replacing more of the system means fewer future failures because it captures the actual source of the problem.

That doesn't mean a chip won't fail. But the parts are soldered to the board just like many, many other parts. I see no evidence that they're the weak link in the life of the logic board and nobody seems able to provide any evidence beyond the anecdotal.

Reliability is the path to reduced ewaste, not modularity. If you want less waste, don't break stuff. When it comes to repairability, there's a clear tradeoff between time, cost, waste. Repair by replacement is often cheaper once transportation and time are accounted for, leads to fewer secondary failures, and a respectable repair shop such as Apple will recycle or refurbish any parts they can.

Likewise, if you want less waste buy what you need upfront and don't plan on incrementally updating it. Every time you update something you have another piece that needs to get shipped by boat, train and truck around the world to your door, the unit it replaces is most often thrown away in the kitchen trash, and you've rolled the dice on triggering a future failure somewhere in the system by tampering with it.

It's worth remembering that these are the parts we're talking about:

1691717490317.png


8 separate chips mounted in 4 different locations around the main logic board.

How would one propose making that SSD modular? 4 separate modules with 4 separate sockets? Making the system thicker only to hold a second layer of board and components? Making the rest of a complex and highly optimized PCB layout less optimal so there is room for an 8 chip socketed rectangle somewhere?

All of those options mean a less optimized primary product that is less reliable, larger, and/or has less battery life.

So there are many, many pro-consumer arguments for soldered components before we need to look for nefarious intent by Apple.



If you are one of the few people who want to open an tinker with your computers, or who actually need to upgrade capabilities after purchase, then this isn't the machine for you. For the majority of people who never update their hardware, it makes for a better overall experience and lower TCO.
 
  • Like
  • Haha
Reactions: throAU and 120FPS
You shared a report showing how much money he likely makes from YouTube and asked me to clarify my view on it, which I did.

If you want me to address Rossmann's arguments in that video, I can't because I won't watch it. He lacks a basic engineering sense and his videos I've seen linked here in the past have been demonstrative nonsense. I won't waste my time or credit him any additional views.



If you want my views on soldered flash chips, I think I've made myself clear but here it is again:

Sockets are a common failure point. They have moving parts. They are assembled by fat fingered humans, not micron accurate machines. The high speed electrical connection is made by pressure, not metallurgy. Corrosion can happen at the interface. They wear under mechanical stress and vibration. If you want a reliable electrical connection solder it, don't push it together.

Touching things breaks them. Every time you open something, touch something, put your finger oils or accumulated static charge on something, you degrade its reliability. Every time you manipulate a socket or connector you wear it and reduce its reliability. Pins bend. Opening things violates seals, if not reassembled correctly you can stress things, breach EMI shielding, breach moisture sealing, fail to ground things properly, loosen screws leading to mechanical wear over time, over tighten screws leading to stress and fatigue, forget to properly power things down, bridge live electrical lines damaging components even if you have powered it down, short a battery pack, etc, etc, etc. If you want to make something reliable, don't touch it and don't make it easy for other people to touch it.

These types of failures are rather insidious and do not show up immediately. They can take months to manifest. Most often a first failure might be a system failure but a second failure is because of the human intervention.

When a part fails, it's very often not the source of that failure. It could indicate a failure elsewhere in the system where replacing a module will just lead to another future failure in that part or another because the underlying problem was not resolved by replacing the module. Replacing more of the system means fewer future failures because it captures the actual source of the problem.

That doesn't mean a chip won't fail. But the parts are soldered to the board just like many, many other parts. I see no evidence that they're the weak link in the life of the logic board and nobody seems able to provide any evidence beyond the anecdotal.

Reliability is the path to reduced ewaste, not modularity. If you want less waste, don't break stuff. When it comes to repairability, there's a clear tradeoff between time, cost, waste. Repair by replacement is often cheaper once transportation and time are accounted for, leads to fewer secondary failures, and a respectable repair shop such as Apple will recycle or refurbish any parts they can.

Likewise, if you want less waste buy what you need upfront and don't plan on incrementally updating it. Every time you update something you have another piece that needs to get shipped by boat, train and truck around the world to your door, the unit it replaces is most often thrown away in the kitchen trash, and you've rolled the dice on triggering a future failure somewhere in the system by tampering with it.

It's worth remembering that these are the parts we're talking about:

View attachment 2244351

8 separate chips mounted in 4 different locations around the main logic board.

How would one propose making that SSD modular? 4 separate modules with 4 separate sockets? Making the system thicker only to hold a second layer of board and components? Making the rest of a complex and highly optimized PCB layout less optimal so there is room for an 8 chip socketed rectangle somewhere?

All of those options mean a less optimized primary product that is less reliable, larger, and/or has less battery life.

So there are many, many pro-consumer arguments for soldered components before we need to look for nefarious intent by Apple.



If you are one of the few people who want to open an tinker with your computers, or who actually need to upgrade capabilities after purchase, then this isn't the machine for you. For the majority of people who never update their hardware, it makes for a better overall experience and lower TCO.

With regards to how one would make this design modular, I understand this on the Air. Frankly, I don't see how they would go about this on the Air (maybe they could find a way, but that's hard to imagine, the Air is already impossibly thin). But on the 14" and 16" pros? There is no reason that they wouldn't be able fit on in there, so I have to disagree with this argument here. I can see how it'd be harder (I'll give you that), but Apple claims it's a security thing, which I disagree with because they allow users to replace the SSDs on the Mac Studio and the Mac Pro (and I don't think they're cutting corners on security or reliability there, so I don't think security or reliability is the real reason).

Furthermore, a socketed SSD doesn't mean a product is more likely to fail. Sockets aren't really the failure point, SSDs themselves are. Maybe if they designed a very bad socket, but it wouldn't be very difficult to design a socket that doesn't fail (and if it did so happen to fail, it would not be very hard for a technician to fix). What is, however, a big point of failure is the dead SSDs that end up with entire Macs being thrown in the garbage bin when they fail because they weren't socketed, and because third party technicians couldn't repair them because they couldn't even source the parts for them.

Now if they went as far as Framework laptops do, that'd probably be another story. I'm sure those do have additional points of failure (and their customers are willing to take on that risk in exchange for the advantages Framework offers them). Should Apple go that route? No, I don't think so. But the idea that Apple soldering them, or locking down the firmware/bios to the SSD NAND chip itself, or preventing the parts from being sourced? I don't think that they necessarily have to do any of these things to attain the same reliability, security, and battery life, and that's the point that I'm making here.
 
It does, when he ventured in to design defects, and developement choices. The responses to his videos are strong because of clickbait/sensational/trollish slant. I said it earlier Apple has poor repairability, i havent bought an apple desktop/workstation in over a decade, because i value ability to upgrade or repair in my workstations. If he had the video with "I give Apple 0/10 for repairability for choosing security encryption and other factors over repairability". I woundt have problems, if not for half truths.

Its funny, he throws around word like defects, try keeping a job in product delivery/developement with that understanding of defect. No company will deliver a product. lol.
You are criticizing his use of the word defect, but I fail to see how you've established that he's actually wrong (other than that you are criticizing non-related things like his use of certain words, or whether he has enough software engineering experience in your eyes). I also asked you before specifically if you could answer the question that I had with regards to whether security and repairability were necessarily mutually exclusive on this particular issue, but I'll rephrase it here:

Apple allows SSDs to be replaced on the mac studio and on the Mac Pro. Are they cutting corners for security (or even reliability) on these products, the most expensive products that they sell? And if not, then how is it that repairability and security are mutually exclusive for their other products, but aren't for their most expensive products where they do allow SSDs to be repaired or replaced?

This is one of the points Rossman (and to an extent, folks like myself also) have been making, and nobody has been able to prove this point wrong. It doesn't matter what Rossman's intentions are, we can speculate on that. But I want to talk about whether he's right. Whether he's using certain words in the English language the right way is completely irrelevant to this point.
 
How would one propose making that SSD modular? 4 separate modules with 4 separate sockets? Making the system thicker only to hold a second layer of board and components? Making the rest of a complex and highly optimized PCB layout less optimal so there is room for an 8 chip socketed rectangle somewhere?

What an absurd argument. So there’s just one way to design a logic board so it will fit into a laptop case? Funny, other manufacturers manage to use fast NVMe user-replaceable storage on their logic boards, at one-tenth the cost of Apple’s storage. Putting the memory on the SoC? OK, I get that. But storage? Nope, that’s just pure Timmy/Apple greed. Apple has been overcharging for memory and storage for more than three decades.
 
You are criticizing his use of the word defect, but I fail to see how you've established that he's actually wrong (other than that you are criticizing non-related things like his use of certain words, or whether he has enough software engineering experience in your eyes). I also asked you before specifically if you could answer the question that I had with regards to whether security and repairability were necessarily mutually exclusive on this particular issue, but I'll rephrase it here:

Apple allows SSDs to be replaced on the mac studio and on the Mac Pro. Are they cutting corners for security (or even reliability) on these products, the most expensive products that they sell? And if not, then how is it that repairability and security are mutually exclusive for their other products, but aren't for their most expensive products where they do allow SSDs to be repaired or replaced?

This is one of the points Rossman (and to an extent, folks like myself also) have been making, and nobody has been able to prove this point wrong. It doesn't matter what Rossman's intentions are, we can speculate on that. But I want to talk about whether he's right. Whether he's using certain words in the English language the right way is completely irrelevant to this point.
I question his basic understanding of engineering, product design choices, what is defective or not. I couldn’t care about Mac Pro and Mac Studio, if you need repairable or upgradable work station, apple isn’t best choice. I use a MBP, and I am sure Apple has more design constraints in a notebook compared to a desktop. It doesn’t give confidence in design expertise talk with lack of basic understanding.
 
  • Like
Reactions: Analog Kid
What an absurd argument. So there’s just one way to design a logic board so it will fit into a laptop case? Funny, other manufacturers manage to use fast NVMe user-replaceable storage on their logic boards, at one-tenth the cost of Apple’s storage. Putting the memory on the SoC? OK, I get that. But storage? Nope, that’s just pure Timmy/Apple greed. Apple has been overcharging for memory and storage for more than three decades.
What other manufacturers have hardware encryption chips in their designs?
 
I question his basic understanding of engineering, product design choices, what is defective or not. I couldn’t care about Mac Pro and Mac Studio, if you need repairable or upgradable work station, apple isn’t best choice. I use a MBP, and I am sure Apple has more design constraints in a notebook compared to a desktop. It doesn’t give confidence in design expertise talk with lack of basic understanding.
You're still attacking his credibility/intelligence without actually disproving his argument. To this point, his actual point still has not been proven wrong.
 
  • Like
Reactions: BigMcGuire
What other manufacturers have hardware encryption chips in their designs?
What does encryption have to do with this?
1) Apple themselves has solved this problem in the Mac Studio and on the Mac Pro, where they do have socketed SSD modules and they are encrypted. There is no reason that encryption is a reason they couldn't also do this in laptops.
2) Samsung actually has a whole business doing this. Apple isn't the only one that does this, if we're going to talk about encryption, other companies have done hardware encryption also.
 
What an absurd argument. So there’s just one way to design a logic board so it will fit into a laptop case?
If that's what you read in what I wrote, then I don't expect more words to be helpful here.

With regards to how one would make this design modular, [...] on the 14" and 16" pros? There is no reason that they wouldn't be able fit on in there
a socketed SSD doesn't mean a product is more likely to fail. Sockets aren't really the failure point, SSDs themselves are. Maybe if they designed a very bad socket, but it wouldn't be very difficult to design a socket that doesn't fail

All I can reply to that is that I'm happy Apple designs their products based on sound engineering principles and not by trying to satisfy the fanciful perceptions users...

Sockets have always been an undesirable tradeoff of reliability for some other goal (modularity, upgradeability, scale in manufacturing, etc). Flat out denying that makes it difficult to discuss it further.
 
What does encryption have to do with this?
1) Apple themselves has solved this problem in the Mac Studio and on the Mac Pro, where they do have socketed SSD modules and they are encrypted. There is no reason that encryption is a reason they couldn't also do this in laptops.
2) Samsung actually has a whole business doing this. Apple isn't the only one that does this, if we're going to talk about encryption, other companies have done hardware encryption also.
That’s your assumption that solution for Mac Pro and mac studio workstations will work in notebooks.
You're still attacking his credibility/intelligence without actually disproving his argument. To this point, his actual point still has not been proven wrong.
I mean this whole thread was about the argument, half truths, clickbait stuff. Lol. And yes credibility and intelligence is important.
 
  • Like
Reactions: Analog Kid
I mean this whole thread was about the argument, half truths, clickbait stuff. Lol. And yes credibility and intelligence is important.
Well, we can talk about credibility, clickbait, and intelligence all day long. But you still haven't actually proven why Louis Rossman is wrong. You've brought questions about his credibility and continued to deflect against the actual argument, but you haven't addressed his actual points.

Also, my point is that security and repairability are not mutually exclusive things. Now there might be other reasons Apple is still soldering their SSDs even into their larger Macbooks, and we can discuss those reasons (Analog Kid has mentioned some things along these lines, and though I still disagree, I can respect where he's coming from because he's citing legitimate points).

But security and reliability are very bad reasons to bring up, and that's the entire point that I'm making. Apple themselves (and many who are on this side of the debate, including many in this thread) have cited security as a leading reason, and I've given my argument as to why I strongly disagree with the notion that security/reliability are reasons for why they must be soldered (despite it being mentioned many, many times throughout this whole thread). I've presented this multiple times and it's been glossed over more than once, and that's the entire point I'm making here.
 
Last edited:
  • Like
Reactions: BigMcGuire
Sockets have always been an undesirable tradeoff of reliability for some other goal (modularity, upgradeability, scale in manufacturing, etc). Flat out denying that makes it difficult to discuss it further.
I still have to disagree with you on this point. I don't think that the trade offs are unachievable for storage, most of the computer industry has been doing this successfully without reliability issues for decades (even in thin and lights). HOWEVER, I can respect your opinion on this and do not want to come across as though I am attacking you in any way. I also will actually agree with you on this point if we're talking specifically about the Air, where these devices are small enough that I think you do still have a valid point (I'm referring more so to the pro, and specifically the larger Pro models where the size constraints are nowhere near as limiting).

We've all been a bit heated on this thread, but I think we can agree to disagree on some things and still see eye to eye on others. Anyway, have a good night. :)
 
Register on MacRumors! This sidebar will go away, and you'll see fewer ads.